Hacker News new | comments | show | ask | jobs | submit login
This is a web page (justinjackson.ca)
1354 points by mijustin on June 20, 2013 | hide | past | web | favorite | 420 comments

First time [1] my in-laws saw "the web" they were trying to find information about vacation spots in Brazil but their travel guide was obsolete, I found a web page a student had written up about where the best places to stay and see were in Brazil.

I could almost see the dots connect when they realized that someone they didn't know, in Brazil, had written up a piece of information at some point that they were now seeing and using, and anyone could do that. It was like watching a Pachinko machine pay out a jackpot :-)

[1] It was circa Christmas 1994 since I was trying to explain to them what Java was and why I thought it might have an impact on the world.

That's a great anecdote. Out of curiosity, how did you find the site? I had just turned four at the time, so I don't quite remember what browsing the Internet was like then.

Because other people have given good answers, I just wanted to add.. the Web was so much better linked together back then. Prior to Altavista made searching useful most of the time, it was common to rely on jumping from one site to another through plentiful link pages, web rings, and lists of resources that people had put together. There was also Usenet which people often posted links on and which was sorted by topic and a handy way to find stuff.


Even after Altavista was clearly better than Yahoo for ad-hoc searches, I still used Yahoo a lot because their topic hierarchy was a really great way to sit down and learn about some topic and its sub-topics.

I was particularly keen on this while eating because you could browse the web with just the mouse very easily.

>There was also Usenet which people often posted links on and which was sorted by topic and a handy way to find stuff.

When my college roommate showed me how to download a graduate student's thesis from Usenet, I thought my head was going to explode I thought it was so cool. Here I am, halfway across the country, and I can download, read and cite some grad student's paper from the University of California Berkeley.

I now feel like a guy who grew up during the 1950's when you tell kids, "Yeah, the internet wasn't always about Reddit and Porn kiddies."

You had alt.religion.kibology and the first crowd-sourcing app in The Internet Oracle.

The technology changes but we seem to keep reinventing the wheel every time the cart gets overcrowded.

Implying we should 'fix' Eternal September is pretty close to implying we should keep important technology out of the hands of those the current users deem unworthy. That prevents both spam and the Arab Spring, both 4chan and Wikileaks.

Keeping the Internet as it was prior to 1991, when commercial access was first allowed, means keeping it a small, controlled entity with obvious choke points that make it trivial to censor or kill entirely. It largely prevents its use for social or political change other than the kinds of change its owners want to see. It would trivialize the Internet by relegating it to a tiny fraction of its current usefulness.

It's easy to idealize what we had. However, losing sight of what we have now, and what we could have, is actively harmful to the prospect of future moral growth.

> Implying we should 'fix' Eternal September is pretty close to implying we should keep important technology out of the hands of those the current users deem unworthy. That prevents both spam and the Arab Spring, both 4chan and Wikileaks.

Not necessarily. The holy grail is growth where culture is preserved, i.e. the rate of acquisition of new users is below the point where they overwhelm the culture.

New users acculturate over time and exposure, so you can roughly model it as "at any time, no more than x% of the users should be 'new'".

That's still exponential growth, so it needn't be elitest. In time, you'll still get to everyone who wants in.

Making it work tho, that's the rub :-)

I completely disagree with your statement. Maybe there is a bigger question: 'Is crossing the chasm to widespread adoption always a good thing?' Maybe we should be trying to build platforms that are high-quality instead of high-quantity. We are caught in a world that is obsessed with inclusion rather than exclusion. Is that, by necessity, the default position? Should it be?

The internet contributed to building support for the Arab Spring throughout the world but it was tyrannical regimes that were the root cause. Wikileaks is not new. We used to have a little thing called 'investigative journalism', a profession in which people would put their live's on the line to reveal scandals and uncover secrets they felt the world should have known.

Technology does not motivate moral growth. People motivate moral growth. Nazi Germany was the most technologically advanced society in the world when they invaded Poland and set the course for World War II.

The reality is that we are segmenting ourselves from other internet users just by taking part in the HN community. Most people here would admit, the minute HN would spread to the point of being widely adopted, we would all leave. I don't come to HN to be inundated with cat pictures. I'm here because I learn things every time I visit a link posted on HN.

> Wikileaks is not new. We used to have a little thing called 'investigative journalism',

We still do. And, historically, it never got us anything as important as cablegate.

> Technology does not motivate moral growth.

Technology enables moral growth by reducing the amount of morality-crushing cruelty we're exposed to on a daily basis and allowing our natural morality to flourish. It's easier to care about others when you haven't just lost your child to smallpox.


> Nazi Germany was the most technologically advanced society in the world

They were, at best, on a par with the rest of the Western World, and, frankly, mismanaging themselves into defeat due to the very nature of their society.

The idolization of the Nazis as being so efficient and so advanced is due to what some Western observers thought before the war, when Hitler was using basic Keynesian-style pump-priming in the form of military spending to get the country out of the Depression, mixed with a few largely worthless wonder-weapons, such as the V2 rockets and the early jets, which did nothing to win the war.

> The reality is that we are segmenting ourselves from other internet users just by taking part in the HN community. Most people here would admit, the minute HN would spread to the point of being widely adopted, we would all leave. I don't come to HN to be inundated with cat pictures. I'm here because I learn things every time I visit a link posted on HN.

This is true but it misses my point: Everyone can have their little niche space, which is amazing if you're not an upper-middle-class heterosexual cissexual white male who would have niche spaces anyway.

Gays can have spaces where being gay isn't weird. Trans people can have spaces where being trans doesn't mean you get your head kicked in. That kind of social interaction was pretty well impossible before the Internet, especially in places like Iran or Alabama, where advertising that you are 'a certain way' could be an invitation to violence.

The Internet is both a network of linked networks and a community of linked communities; both aspects are vital to what it is and what it can become.

I don't know - you can try to be inclusive and also set some standards. That doesn't have to be for the entire Internet.

The future people will have fun comparing and contrasting different reputation systems and online communities. Assuming enough information stays for them to study.

Usenet was archived by Deja, which got bought by Google. It's in a sorry state, but at least it's stil there.

Who's archiving Reddit, or HN, or 4chan, or all the rest?

I agree with your argument. But I think the internet's strongest trait is not necessarily revolutionary changes (wikileaks and arab spring). The things that have been built on the internet have been incredible. Think of crowdsourced projects like wikipedia and how ridiculously trivial it is for a scientist (or someone like me; a student in a lab) to access information or search for experimental data (genes are the tip of the iceberg here).

I don't want to belittle the Arab Spring, but the NCBI database and Wikipedia have arguably had an even larger impact on society than social networking's capability for kickstarting revolution. It's just slower and less obvious.

I was at a computer camp sometime around '94, and the camp had a printed "Directory of web pages" which was exactly like the yellow pages of the phone book. You would look up a subject, and it had a printed list of urls on that subject.

I still have a copy of "The Whole Internet User's Guide and Catalog" published by O'Reilly in 1992. It has a whole section devoted to lists of addresses of sites grouped by subject. Many of the sites didn't have urls because they were operating over ftp or telnet or WAIS or Gopher.

I just checked and the terms "google", "altavista", and "search engine" do not appear anywhere in the index. Also, no weird web2.0-style domain names.

Different times.

Yes! I will gladly second that.

The sheer amount of information back then was nothing compared to what it was today, but it was so much easier to browse around. Something as trivial as a webring was a cement for a community rather than a pledge for traffic.

Despite the nostalgia, I do think that, overall, the WWW experience has mostly gotten better since then, if only for the amount of information and communication. However, the density of quality has rather shrunk...

As late as 1998 I remember a friend griping about people who said that a site on topic X was good "because of the links".

Couldn't the same be said about Hacker News? Search has improved, but we still rely on "sites with links" to find interesting content.

He (and they) meant the static links, not the regular updates.

ah.. web rings.. I remember babylon 5 web rings... Nostalgia's a hell of a thing.

If we knew back then what is common knowledge today, we could build an early version of Google search in a few months. Web spider, parsing HTML, extracting links and text, computing Page Rank and then making an inverted index to look up info by keyword. Imagine this would have been 100 times better than Yahoo's links and Altavista's search.

By the way, syntax highlight thinks "Altavista" is a spelling error. Talk about becoming the footnote of history. Google was the company that took the lead from AV and now Google's own browser can't understand their main competitor of 10 years ago. AV did it to themselves, putting all that shit on the homepage and not providing good results. LOL

Search was actually anticipated (along with social graphs) by TBL in his first WWW proposal documents, posted to HN a few months back.

It was anticipated in the protocol, that's what ? is for.

Web Rings were amazing though; found a great site about X-Files plot theories? -- no problem, chances are you'd find more that were like that by following the web ring. We lost something back then that, even today, we never really got back...

Does anyone know why altavista.com and altavista.digital.com were different? I just remember preferring the digital.com version for some reason.

I wish I could go back in time to 1996... everything on the internet seemed so much more meaningful, not to mention textual (because images took so long to download).

> everything on the internet seemed so much more meaningful

Back then it almost felt like we were in the heroic age of internet communities, to borrow a phrase used about a period of American inventors. It'll never be like that again, but there are more good things coming in the future.

So, the web back then is what Wikipedia is today?

That's not a bad characterization.

Web pages were static, many were long-form, they tended to provide information, and link copiously to one another.

What they lacked was collaborative editing, though Ward Cunningham was already working on that at the time. Wikis postdated the Web only by a brief bit.

Wow, web rings, huge nostalgia for that!

Did you use tabs back then?

I wish there was still a curated list of categorized websites.

Back in the day you bought magazines like ".NET" (a UK publication) that told you where the hotspots were on the web.

There were other physical publications like the "Internet White Pages" or some such thing (no seriously) which I remember seeing in UK bookshops and thinking "that's gonna be outa date in a week or two".

Back then you felt you could contain all of the websites (or pointer sites) in your head. It was truly an exciting and exploratory time.

Oh and I forgot, using KA9Q [1] for your TCP/IP stack on Windows 3.1, because Microsoft had't quite caught up at that point. And hacking the UK's PSS network to find your way into a university that was connected so you could access usenet. And the early days of Demon Internet in the UK....oh the nostalgia of it all :)

[1]: http://www.ka9q.net/code/ka9qnos/

Ah yes, you are probably referring to the Internet Yellow Pages[1], use to be one of my favourite things to use at a library and write down websites..


That one as well. But there was also "The 1994 Internet White Pages" published by Wiley Press which is credited with Seth Godin as a co-author [1][2].

[1]: http://i.imgur.com/HxDWs9c.jpg [2]: http://amzn.com/1568843003

In 1994, the main way you found websites was mostly from hub sites that maintained large libraries of links to other sites (like a directory), newsgroups (UUCP/Usenet), friends in IRC trading links, AOL, Delphi and Compuserve forums and print magazines. Webcrawler was revolutionary when it came on-line in 1994, but there were no comprehensive search engines that had a high index count until probably 1996.

The mother of all link directories started from NCSA who distributed the first web browser (Mosaic) and they ran a site where they would publish a monthly list of "What's New" on the web. They pulled the pages from their site a few years ago but fortunately I made a mirror on my server for historical reference that you can check out here:



edit: forgot to mention web rings someone else just mentioned.

I remember circa-1994/1995, our library got a book called something like "World Wide Web Yellow Pages" and it was a printed book of a couple hundred pages of curated web URLs. Even then the irony was not lost on me. Why didn't they just put it online? The answer was, well, then how would you find that?

Many sites, including most home pages (remember those?) had a page of links, a small curated set of links[1]. Ooh, and guestbooks! Thousands and thousands of guestbooks!

[1] http://web.archive.org/web/19991022011955/http://www.geociti...

Mosaic, totally remember those WorldWideWeb days.

In context, I was part of the group that would eventually become the Java group (at that time it had transitioned from being the Oak group to being the 'Liveoak' group[1]) Patrick Chan had written a web spider in Java, we marvelled that it seemed like it could visit every site on the web over night :-) But mostly what it found were the 'big' web sites (computer companies, early movers in the Internet space) and students who had downloaded X-mosaic and the httpd from CERN and were playing around with early HTML.

[1] The name was a somewhat morbid joke on James Gosling's part as Sun was cancelling the Java project at the end of the fiscal year and had been split into the "part that the hardware guys liked" (Sun Interactive) and "the part no one knew what to do with." (Liveoak)

I don't know about the OP, but in 1994, I was mostly using Webcrawler as my search engine. Lycos came out around then too. Switched to Altavista pretty quickly after it came out.

Quick answer: links. Hypertext. It's important enough to the original vision of this whole contraption to actually be the name.

While were reminiscing... In 1994, the web felt like technology. You typed in "http://" and it felt exotic. You were connecting to a server. Literally: _a_ server. You wondered what would happen if you hit the same address with gopher or (more likely) telnet.

As often as not, something would happen.

I didn't have a computer at the time, but my friends who were online all used AOL, which came with its own search engine, chat rooms, email, etc.

One year away from being annoyed at the web to being blown away by yahoo.com. A friend searched for some game or something on yahoo and I was like what is the trickery?

I feel like such a hipster for having used AltaVista and Dogpile as search engines.

What is hipster about that? Your statement makes no sense.

I'm 46 and started using the "internet" in 1992 - usenet, gopher, FTP et al. Then came the World Wide Web". Altavista was really all there was before google, why is that "hipster"?

That's like saying you were a hipster for using horse and carriage before Henry Ford started mass producing cars.

If anything being an early adopter of Google seems more "hipster" than using Altavista. Gawd, I musta been and early hipster, how much more "hipster" can you get than that.

Between Altavista and Google, alltheweb.com (FAST Search) also had their day. I was in charge of a major web portal around 2001, and we were contracting out site search queries to FAST. One day I got a call from a new company named Google. The guy said he could give me a better price than FAST. I asked what price. He said, "Any price."

Little nugget:

the FAST Search guys sold alltheweb.com to yahoo.com and focused on enterprise search (i.e. they sell/license their software to companies) with customers such as newspaper companies, yellow pages companies and government agencies. They were then acquired by Microsoft a few years ago and rolled into their Sharepoint offerings.

Dont forget northern light. It was a great search engine targeted at researchers and academics that grouped results in folders of topic areas. They made the unfortunate business decision to transition to a paid service right as google was becoming popular.

lol that's great. Wonderful to think that even Larry and Sergey had to hustle. Everyone has to put in their time.

To have used Altavista or Dogpile one must have either been on the internet before it was cool, or made a deliberately retro choice of search engine. Either qualifies for hipsterdom.

And Astalavista.

yep, astalavista.box.sk, still remember the full domain by heart... :)

Edit: OMG it's still alive..

Of course it's alive. Check the whois data for that IP, hosted by "ANTIK Telecom". :-)

code.box.sk was my favorite resource on programming languages then. The whole *.box.sk was amazing.

My god, it even uses tables for the layout, takes you back uh?

Check out hacker news.

Hipster huh? Well what about Lycos, remember that search engine?

Remember the search aggregators, like Magellan (maybe Columbus? Something like that) that searched X different engines and aggregated the results in a desktop app for you to use?

Then I found Google, and I never used anything else again.

That was Copernic. (After Copernicus) I remember using it and there was a list of perhaps 20 engines. I remember the first to return (like, would return in 1 second not 30 seconds for the rest) was google.

Every single time, google was the first back. It wasn't always ranked the top result back then, but it was fast.

Yes! That's it. Yeah, Google came and blew everything else out of the water. I remember searching for an ocx file for something that wouldn't run, and everything kept returning some spam sites. I somehow stumbled upon Google, after some recommendation, and tried it, and the first result was a legitimate download of the actual file.

It's not hard to see why everyone switched to it.

That's what Dogpile did - at one point - albeit as a regular browser search engine, which was its allure to me. I often got better results than on Google, which must have been around their formative stages.

HotBot was where it was at.

Oh yeah. Vividly.

Just like your Pachinko reference

Just being cynical here, but books written by people from far away places have circulated for a lot longer than the internet has been around.

Not just anybody. You had to have a publisher, and it took weeks to years to get a piece published. Which means nothing got published if there wasn't a chance at a monetary return. The web is a significant improvement if you ask me.

Before Gutenberg et al, you had to have someone spend weeks or months to hand copy your book, one at a time.

To me, the web is just the next step in the improvement in the publishing of the written word. I don't see it as different than a book.

First time my in-laws saw "the book" they were trying to find information about vacation spots in Brazil but their grandpa's diaries were obsolete, I found a printed book a monk had written up about where the best places to stay and see were in Brazil.

I could almost see the dots connect when they realized that someone they didn't know, in Brazil, had written up a piece of information at some point that they were now seeing and using, and many people could do that. It was like looking into the darkness and suddenly seeing a firefly blinking :-)


My point here is, the printing press had it's impact, nobody denies that. Seeing the times changing is what leaved ChuckMcM's in-laws astounded.

"At it's heart, web design should be about words."


The web is not just a place for text / words, that is not its heart and soul, and therefore neither is that the case for web design. It really never has been. I see no great argument in favor of words being the core over any other form of expression. Today's bandwidth more than allows for beautiful video, animation, high quality graphics and photos, etc. Or low key graphics, subtle interactivity, and so on. It's absurd to argue that such a rich medium should always be focused on words.

Text does not have to be the focus. It depends on what the purpose to be achieved is.

And yet you chose to present your ideas using words. At the heart of the Web is the communication between humans. At the heart of that communication are words.

You can argue that we can communicate using images, but that communication is limited when compared to words.

You can argue that we can communicate using sounds, but unless those sounds are words, the communication is limited in the same way as when using pure images.

You can argue that the Web is not only about communication between humans (e.g. we can do shopping online), but the communication is indispensable for all those other services.

You can argue that the Web is not only about communication between humans, but all that machine-to-machine communication is ultimately there to connect humans.

Web design is there to help you achieve whatever you're trying to do on the Web in a way that's more appealing and easier to interact with. I think that Justin's assertion is that words are at the heart of whatever you're trying to do on the Web. And I think he's dead on.

> You can argue that we can communicate using images, but that communication is limited when compared to words.

Like so many other things, it depends. When people are online looking to book a bed and breakfast, buy a pair of shoes or find a date, they tend to find images very valuable indeed. In fact, in the absence of images, many of the web's transactions would simply not happen.

It's not for nothing that we often say, 'a picture is worth a thousand words'.

It's funny but this is not really true. We've just finished a massive UI test involving a choice our users have to make during sign up.

Various graphical images illustrating the choice caused more confusion than a short sentence for each.

We also found replacing text links with icons increased support requests from new users by a noticeable margin. AND when keeping the icons and restoring the text it was still above baseline

My view is that for length and complex ideas images or video can be illustrative. But for 90% of interactions text is always clearer.

Your example doesn't invalidate my point at all because it's restricted to one particular situation: signing up for something. If you were selling shoes online instead, I think you'd be much more likely to agree with me on the importance of images, which would simply prove my point that "it depends".

Consider what works better:

* a description of a place and how to get there or a map

* a description of a floor plan or a diagram

* a description of someone's appearance or their photograph

* a description of a piece of clothing or its photograph

The OP said, "the most powerful tool on the web is still words". That might be true, but I disagree with CodeMage when s/he said that "words are at the heart of whatever you're trying to do on the Web". I don't believe that words are at the heart of Google Maps or Flickr or last.fm or countless other web properties that revolve around sound, images and video.

If pornography were wordless, would people stop watching it? Do people watch it now because of "communication"?

Words are important, but they are not of sole importance in general and in many cases are not of primary importance.

"It's not for nothing that we often say, 'a picture is worth a thousand words'."

This sentence can support either side of the argument. I love it. Draw that sentence.

I don't think anyone is completely devaluing images in communicating things; they just assign images to a lower priority in the communication order, one that is supportive rather than primary. Without the context of buying shoes, pictures of shoes tell me nothing. I can fill those shoes with all sorts of stories.

(Static) pictures usually don't have implied ordering of information, unlike words, and videos, which require serialized focus to perceive.

One word (pun intended): Hieroglyphs.

It's unfortunate that the article doesn't try to actually make the case for the primacy of linguistic communication, and I'm not going to take the time to do it here either. I'll just say that while I agree with you that a diversity of media is better than just one medium, I spent my childhood, before the Web was built, reading books instead of watching TV on purpose, not by accident. Books without pictures. And it's not an accident that I'm answering you in text instead of using my webcam.

I have occasionally seen beautiful video on the web. I have much more frequently seen video whose author had deluded himself into thinking it was beautiful.

http://lists.canonical.org/pipermail/kragen-tol/2013-June/00... talks some about the relative usefulness of different kinds of digital media.

Do you have a good citation for the primacy of linguistic communication?

I have become profoundly convinced over the last 15 years that words (& to an extent, text), is the generally best mode of communication of meaning, beating out things such as video, images, smells, etc. But I've been generally unable to articulate that idea cleanly and persuasively.

I feel that words are the best way to communicate because words are born purely out of us; they are literally our thoughts in physical form. Videos, music, pictures, are all representations of physical things in the world around us; words are representations of the worlds inside us.

I sometimes have a hard time putting certain thoughts or feelings in to words, in which case I either draw a picture, play a melody, or use one of the other myriad forms of expression available to me to communicate with others. I in no way feel like I am translating these thoughts from words to these other forms of expression. I have a much easier time singing a melody than transcribing it. I have to force the music to comply with written forms.

> Videos, music, pictures, are all representations of physical things in the world around us; words are representations of the worlds inside us.

What about music, sounds?

Al Gore articulates the value of text well in _The Assault on Reason_. Short version: text is a "cool" medium which only indirectly rouses emotion, while images, sounds and especially video are "hot" mediums that are much more likely to bypass the prefrontal cortex, and directly stimulate emotional reactions.

That sounds like it derives from McLuhan's theories in "Understanding Media: Extensions of Man", a book I'm currently reading.

Ah, you are correct! Gore's book is merely where I learned it from.

I'm not sure it is universal. I love the written word [1]. You can be precise and make nuanced arguments with caveats and qualifications where required. You can skim when reading it in a way you can't[2] with video. However some things are better and more easily shown than explained and some people have a preference for visual explanations.

[1] As I suspect most people spending much time on this site do as they aren't here for the pictures.

[2] Actually video editing software does give mechanisms for rapidly scanning the video so perhaps it is possible but not offered in web video software that I have seen.

Maybe you should try expressing it with a picture :P

A good citation? It's a value judgment.

> I have much more frequently seen video whose author had deluded himself into thinking it was beautiful.

I have some sympathy for your point, but I don't think the argument about video is valid. 99% of the web is crap. Always has been. Probably always will be. The crap includes text, video, etc., etc. That's why we have search engines and sites like HN -- to help us find the 1%. The existence of the other crappy 99% says pretty much nothing about the relative advantages of text vs. video.

Point — but you're even less optimistic than Gresham!

I meant Sturgeon.

>The web is not just a place for text/words, that is not its heart and soul... It really never has been.

I am thinking back when they named the protocol the HyperText Transport Protocol and the primary format the HyperText Markup Language, text was indeed the heart and soul of the medium.

At one point, images weren't even displayed inline but via a helper app. This is why Mosaic was a big deal.

Yeah, that's all fine and dandy, but what if we measure "what the web is for" in bandwidth? What do most people use the web "for"? Netflix and YouTube, I'd wager.

The heart of most TV Shows and Movies? A screenplay. Words. Dressed up with faces and costumes and animation, but words nonetheless.

I'll grant you that YouTube has a lot of cat videos, which probably doesn't have words in it. But those vlogs have a hell of a lot of words in them too.


A lot of video is streamed with RTMP, yes, but a lot of Netflix and YouTube content is available over HTTP now too (albeit in the contrived manner of chopping videos up into dozens of tiny files).

Words put to effect with the breadth of human expression captured through full motion video. I challenge one to read the screenplay for Shawshank and feel half as much.

What the author extols in word is something quite effectively conveyed in many forms of art. Just as there is great beauty in the text of Baudelaire, there is incredible emotive power in koyaanisqatsi.

We should accept the author's challenge to behave with more conviction and purpose, to express ourselves in a more meaningful way. But to do so in word alone is too restrictive for the imagination to bear.

> I challenge one to read the screenplay for Shawshank and feel half as much.

It is adapted from a Stephen King novel, having read it I can tell you you feel just as much.

No, you are thinking of TCP. This is an article about HTTP. (in terms of bandwidth, HTTP ⊆ TCP)

Pretty sure YouTube is transported over HTTP. Not sure about Netflix but I wouldn't be surprised.

HTTP has become a generic TCP wrapper for the common request->response pattern.

Wireshark doesn't think so:

9792 42.750939000 TCP 1506 [TCP segment of a reassembled PDU]

Dunno, I tried loading a YouTube page and capturing packets. Only ports in use were 443 and 80, even to the cdn-like hostnames (bla.c.youtube.com). First packet shows "GET /videoplayback?...."

A port number is just a port number, it doesn't define what protocol is in use. You can serve different protocols across it - the limitation is that generally you can only have one service listening on a port number. A youtube page will require a couple of different protocols.

Youtube uses RTMP on port 80.

RTMPT is RTMP over HTTP, but Youtube doesn't use it.


It is not transported over HTTP, although it does make use of HTTP at the Application Layer. It is transported over TCP, which is a Transport Layer protocol.

Yes, I'm being pedantic and annyoing.

People are using HTTP as a transport layer, e.g., websockets.

Websockets are the transport layer, built over TCP. Websockets happens to use HTTP as a protocol. That doesn't mean HTTP is the transport layer.

Porn and cat pictures.

The web is not just a place for text / words, that is not its heart and soul, and therefore neither is that the case for web design.

I'd say that text is not only the heart and soul of the web, but also of motherfucking civilization. Animation and video are nice, and great adjuncts to text. But language is still the most common and easy way to store and share information intended to be consumed by humans, and since that's what the Web is all about, it's not only natural that the Web should be language-focused, but also unnatural that it be anything else.

Isn't there a difference between text and language? I see a lot of these comments getting them confused.

Drawings are the heart and soul of civilization. For one, they originated ~37 thousand years before the first written language. Second, understanding text requires knowing how the information it contains is encoded (what language it is written in). Drawings are far more universal.

Perhaps, but Classical Greek civilization remains startlingly intelligible to us after 2500 years. We stare up at the paintings on the caves at Lascaux having only the vaguest idea what they mean (and that's hardly the most recent example).

One might also imagine how our understanding of, say, the Italian Renaissance might be affected were every painting to disappear, and then to reconsider the same scenario with every bit of writing erased. Both would be devastating, but as a student of human culture, I'd far prefer the former.

The fact that images pre-date anything vaguely similar to what we call civilisation shows that the two are not causally linked. Words /are/ causally linked with civilisation.

It is speech that is most universal -- words. Drawing came after, and relegated to being foci of rituals and stories that were spoken and symbolic of spoken agreements.

And I read picture books before I graduated to books with text and chapters.

  > Second, understanding text requires knowing how the 
  > information it contains is encoded (what language it
  > is written in). Drawings are far more universal.
So you're saying that you can look at those old drawings and understand what they mean?

'language' != 'text'. When you're having an intimate moment with your significant other, do you use your voice to convey your love, or do you write it down on a bit of paper?

I agree. Also, upvoted for motherfucking civilization.

Words delivered as electronic text are a great equalizer, for both the writer and the reader. Words impose no accessibility barriers for people with sensory disabilities. Words require minimal bandwidth. Even for readers with no significant disability or bandwidth limitation, words are great because the reader can read them in whatever way he or she sees fit (computer screen, e-ink display, printed page, text-to-speech).

Written words allow a writer to communicate without being hindered or prejudged by things beyond his or her control, such as accent, rate of speech, sound of voice, or poor acoustics. A writer can rewrite and edit his or her words to say exactly what he or she wants, far more easily than one can edit a video or audio recording.

Electronic text is more amenable to analysis and manipulation (e.g. translation) by machine than any other human communication medium.

I'm not so sure about this, but written words might favor reason over emotion, especially if one can disable presentation hints.

For these reasons and more, the Web should be primarily about written words delivered as electronic text.

>>"At it's heart, web design should be about words."


There's some strange dissonance going on in my head reading this on a site with practically no images at all.

Once images went inline, the web's focus moved from content to presentation, much to its detriment (in my humble opinion). We progressively became more at the mercy of the page designer rather than the copy editor.

I don't think posting a comment counts as "web design". Nobody said "writing comments shouldn't be about words".

There is a "design" to this page. It just doesn't include pictures, videos, hover bars, animated banners, nor advertisements for weird tricks to lose weight. This site is about the flow of textual information.

I like a usable application interface as much as the next guy, and am quite happy to see native interfaces (even if they aren't my beloved WPF ;-) rise back into prominence so that the hyper-text transfer protocol can get back to carrying hyper-text.

The people who built HN did the web design you speak of; Web design is not what users posting comments are doing.

Yes (except I am not speaking, I'm whispering).

I never implied that posters and commentators were doing web design while using this site; I started with a comment on the dissonance of reading a statement refuting the web as a textual medium on a site that is effectively only a textual medium.

On the other hand, yesterday there was a discussion on HN of Steve Jobs remark that his legacy won't survive long, because technology quickly becomes obsolete:


Words have much better chance of survival than other elements of web design. What was considered a decent design 10+ years ago, is today often laughable.

This opens a question on what exactly elements of web design are of course. Some would claim content isn't an element of design, others would claim the opposite.

I would just remark that photos and video have the same survival aspect to them that text does. Particularly as our present and future is clearly one stored digitally. It's all bytes.

I would also point out for fun, that duration of survival doesn't inherently equate to greater value or beauty. There are some ugly and boring things that last a very long time (rocks vs flowers; I think I can say that people tend to think certain flowers (roses) are inherently more beautiful than your average gray rock).

Sites made ten years ago with a design that focuses on the text still look good. Sites with a more "visual" design, even one that was very well regarded at the time, look far more dated.

not technology but tool. apple // was tool. It is obsolete. Netscape was a tool. Obsolete. Is TCP/IP obsolete?

The trend of today is sticky headers. Tomorrow it will look rediculuous. Is it tech? No. It's the matter of fashion.

Technology is tools.

"At it's heart, web design should be about words."


I corrupt too many hours concerned with flash, and dressing words in garish nonsense. Minimalist HN is an excellent model to excise and distain word flash dreck, and images or whatever kind of vid snippet.

Look at HN's traffic, others agree.

With the myriad digressions of image, js, and html5 magic, it's difficult for me not to be seduced by other's new eye candy. But the best pages I like to find, are simplest original tty, possibly with a word centric CGI engine dynamically serving up what I want, or what I didn't know I want. Most often then not, these plain forged dynamic/static pages, I bookmark.

Some site's landing page ar so busy loading, to read what I want I drag-copy or 'print' if that's an option. If not, forget it, it's just too much painful nonsense spam or Las Vegas sequins headache to tolerate.

I'd also prefer a text browser, but gave up as none seem to keep up with required js, certificates-ssl, etc.

I'm no longer an infant, or a girl believing in fuschia unicorns, let's lose those brain-numbing colors too.

I honestly believe that we could fix HTML by stripping javascript out, or narrowing its scope such that it was incapable of asking anything about the presentation of content and interface devices.

Why should anyone care whether a web-page is being read by a human using a cell-phone, from a terminal, by a blind person with a braile reader, an elderly man with essential tremor (hint: fiddly pop-up-and-disappear menus are really hard for some people), someone using JAWS or other screen reading technology, whether someone is actually not a someone at all, but a rather a computer program Jim wrote to alert him of news stories he might want to read. Why should we care if someone uses a mouse with or without a scroll-wheel, or if he has an 24-bit display?

It's frightening to think that the peak of web-accessibility may have been in the past, and that the web is getting harder, rather than easy to 'design for', as people slowly but surely continue to try and gobble up the browser's job of presenting content, so that they can get it to do display according to their own arbitrary sensibilities instead of expressing basic intent.

> I see no great argument in favour of words being the core over any other form of expression.

I think the article zooms in on text for "shock value", if you will, but there's a deeper point.

I trust you're not on Hacker News because of the aesthetically pleasing orange bar at the top of the page, you're here for the news, and the discussion. And that's what the article is getting at.

We get so wrapped up in the tech stack powering our websites, and oh so elaborate presentation layers, that we sometimes forget that what truly, really matters is the content encased in all that.

Right, and there's a very strong argument to be made for a text focus on Hacker News. As I said, it depends on what the purpose of a site / web page / etc. is.

The point was that words are not the heart and soul of web design. There's nothing that argues in favor of an inherent bias to words over other forms of expression. That's my position.

Of course words are more important.

Imagine the web without images. Maybe Boring, but largely functional. Now imagine the web without words.

I see no great argument in favor of words being the core of any other form of expression.

Except the novel, the poem, the play, the scientific paper, the biography, the corporate operating agreement, the rule of law...

I think the authors point is that good content is more important than form and not that graphics and media should be forbidden. It just happens that HN is an excellent example for that - people come here for the content and not for fancy design, graphics and media.

A friend of mine has a special weakness for Venice (Italy) because it is one of the few place without any billboards, i.e. graphics everywhere.

It is not black and white but good food for thought.

I think his point is that it's a communication system that facilitates communication amongst everyone. And everyone can communicate in words, typed out[0]. And not everyone can produce or effectively communicate using beautiful video, high quality graphics and, I guess, responsive design. [1]

It's interesting to wonder what the next-common "wide area" communication mode is -- photos are widely shared, but is there a lot truly communicated with them? I guess it depends on whether you think those picture's typically say more than a thousand words. Videos, too. They serve as a topic of discussion amongst peers, not as a mode of discussion themselves.

In fact the deeper point remains even if I conceded your argument: there are always those who excel in the creation of art and in presentation, regardless of medium. What makes the Internet different is that it facilitates peer to peer communication, that quality powered for the most part by the universality of written text. I guess that's why it if often compared to the invention of the printing press, another time such a fundamental shift in human communication took place (facilitating many to everyone, but not everyone to everyone).

Those who excel in arts and communication -- and, to an unusual degree! engineering -- are still the magic wizards/grey eminences of the medium, prominently shaping something that wouldn't exist without them.

[0] Words, typed out: A manner that's arguably truly more efficient and useful than what it's originated to represent, speech. Imagine consuming all the text content you read everyday on the web (such as this!) delivered by speech.

[1] Leaving out the fact, that of course, not everyone has the ability to participate in words, typed out. While I do think it has served to interconnect everyone in terms of exchanging thoughts, it also serves as a means of separation between those who can and those who can't: lacking access or straight up capability to use this mode.

Edit: added some more thoughts.

"not everyone can produce [...] responsive design"

Notice that the page is automatically responsive, and is perfectly usable on a phone, tablet, or anything else.

If your page has no CSS styles it will be perfectly responsive. If it's not, you've messed up that nice feature of the web with your stylesheet. Now you probably think you need to learn about media queries to make it work again.

I would actually argue that raw text is not 'responsive,' because column widths that are as wide as a large monitor are incredibly difficult to read.

I've never understood this attitude. If your monitor is too wide for comfortable reading, don't maximize your browser. Why would you ever want a window that was 2/3 white margins? Site makers: please please please let your text expand to the width of my browser. I made my browser that wide because I like it that way.

That's a good point, but you are sort of redefining responsive. The unstyled HTML (not raw text) will still flow to any size window. The problem you mention can be fixed with a "max width", as in the original article.

I'm not sure that there's a universally agreed upon definition, but I think my comment certainly is relevant if we go by the one on Wikipedia: "Responsive web design (RWD) is a web design approach aimed at crafting sites to provide an optimal viewing experience—easy reading and navigation with a minimum of resizing, panning, and scrolling—across a wide range of devices (from desktop computer monitors to mobile phones)."

Everyone can communicate in words? That isn't true, unfortunately. I don't see why you even said that, only to negate it at the bottom.

More people can digest a photo than can read a specific language, by a radical margin.

I agree that an individual pictures or movie has the potential to be understood by a larger number of people. Even by a "radical" margin. :) However, that's not a unique feature of the Internet, it's similar in all means of communication. In media older than the Internet, pictures and movies have had an impact across linguistic communities.

But the vast majority of pictures and movies aren't the My Lai photographs or the moon landing. Most pictures and movies just don't communicate a whole lot of thought, it's amazing when they do, which is why photography and movie-making and music are such arts! That's because the vast majority of everyone doesn't have the capability, opportunity or sheer (in fail videos: bad) luck to take a picture or a movie like that.

In other words, visual arts might be a more universally accessible when it comes to receiving side, but they are much less universally accessible when it comes to the producing side.

All of your contributions seem to denote someone who doesn't really have anything worthwhile to convey in writing. Try to communicate "Hire me, I'm experienced." in a photo. Share that on Facebook.

If you're going to attempt to insult me personally (instead of focusing on the content of my posts), in words, I'd prefer that you put in more effort if we're talking about things that are worthwhile.

> All of your contributions seem to denote someone who doesn't really have anything worthwhile to convey in writing.

Please refrain from this sort of ad hominem response, it adds nothing to conversation.

> And everyone can communicate in words, typed out

Obviously false, or else answer me in Cherokee, Tamil, Hindi, and Latin.

Images can be understood beyond the bounds of a single language.

Likewise false, or else find me an image that very clearly indicates the same content as, say, the Gettysburg Address.

Images can be understood beyond the bounds of a single language, but only for the very simplest of concepts. Go more complex than a sentence or so, and interpretations begins to muddy the waters in ways that language simply does not.

Images can be perceived and a narrative constructed regardless of language, but this may not be the same narrative the creator intended, in which case, nothing was communicated.

Those languages use words.

How many of those languages contain words you understand?

Written language is a system of images — sometimes with phonetic significance, sometimes ideogramatic.

While your argument seems persuading at first read, a second glance makes it clear that it's a refute just for the sake of it.

In defense of words, hypertext markup language has a lot of tags for words (more than I could count on two hands) vs two hands worth for images, pictures, video, etc combined. Throw in CSS (which makes words pretty) and you have a lot of stuff to do with words, and not that much to do with arrays of pixels or sine waves.

Ok, but I'm not sure how the quantity of tags determines the value of the content contained.

I can vary a graphic element or photo a near infinite number of ways prior to inserting it via one tag. I wouldn't say that argues in favor of imagery over text, or vice versa, however.

I'm not arguing the value of the content, just the original intent. HTML was just meant to be a markup document language to display rich text over http.

I definitely get you on that. Does that original intent say anything of whether words should be the heart and soul of web design however?

Should black and white photography be the heart and soul of photography just because limitations (or costs) of technology started us there? I'm guessing you wouldn't argue that original intent should define future value, but these are all interesting points to discuss.

This argument really irritates me. Apart from all of the excellent reasons that other commenters have given as to why this is false, the fact that programs such as Instagram are as popular as they are, /that do nothing other than remove everything but the text from a page/, shows this is false.

It's funny... there haven't been many photo responses in this HN thread. ;)

Agreed. At it's heart, web design should be about conveying information effectively.

Agreed. The internet is essentially about information. Words are a subset of information transmission. What we can take away from this page though, is that information should not be obstructed by cruft. I agree with that.

Agreed. The author is assuming our capacity for communication is/should be text based. We're visual beings as well, and aesthetics is a lubricant for effective communication.

Absolutely. I clearly should have posted this on Instagram. ;)

change "words" to "content".

I don't think so. That isn't what the person said. Their entire post was built around stripping down web design to the words. In fact, the author was very precise in the usage of the term "words."

This for example: "But the most powerful tool on the web is still words."

That blatantly is not the case in my opinion. Photos are vastly more powerful than words on today's web. That's why 3/4 of all the posts on Reddit's home page are links to imgur. That's why a gazillion photos are uploaded to and shared on Facebook every day.

That's why you can read about protests in Brazil all day long, and until you see the photos, the sheer power of the protest will not ring anywhere near as loudly. This has been true so long as photography has existed. Whether we're talking about the Holocaust and WW2, an atomic bomb going off, or a photo of a kitten or puppy, you name it and photos win 9 times out of 10 over words.

What would walking on the moon be like without photos (and video) to go with it? Would reading the transcripts have amazed the world like the imagery did? I say not even remotely. Imagine reading about a man standing in front of tanks in Tiananmen Square, versus seeing it.

Another side note to all of this: there are almost a billion illiterate people still, sadly. And another billion that are only partially literate. Nearly every one of them can understand imagery. Imagery also crosses languages, you don't have to understand Mandarin to grasp the Tiananmen Square protest.

I have never seen a still image that conveyed the horror of the Holocaust to me in the way that Elie Wiesel's Night did. I have seen a movie that did: Schindler's List, which was fiction, made with actors; and which was made at enormous expense because of the things people had said and written.

Beyond that, though, while images and movies are highly effective, if very expensive, ways of influencing people's feelings, they work just as well in the service of falsehood as in the service of truth. As such, they primarily work to amplify the influence of the powerful. Language allows you to share your thoughts instead of your feelings, and text in particular provides a much better way to gradually explore our way toward the truth. With language and text, mere power stands helpless before reason.

OP here. My point wasn't that we should remove all other types of media from the web. (I love other types of media)

My point was that the most powerful tool on the web is still words.

Here's why:

1. Words are an excellent starting point. All you need is a text editor. Words are a great way to express ideas in a concrete way. Words can stand on their own.

2. It's easy to add things to words. Adding an image, video, or other media to words has the power to further strengthen the message. Side note: there's also the (negative) potential for adding too much, or for muddling the message.

3. Text on the web is small. My page was 4KB. Most people around the world, even with a bad connection, would be able to view that page.

You're right: images can be powerful. But on the web, I think text still rules.

(I notice you responded to this thread with words, and not an image or a video)

I would go one step further, and say that while images are very popular on the internet, most are light on meaning.

The parent mentions the some large percentage of the links are to imgur, which is absolutely true. However, the actual content of those images is so shallow they could each be summed up in a sentence apiece; ironically enough, looking at the front page this very instant, half of the images contain words to convey a specific message, and the rest rely on the title of the post to convey context to the image.

The right picture may be worth a thousand words, but much to my dismay there are very few pictures of that quality to be found.

> That's why 3/4 of all the posts on Reddit's home page are links to imgur

They still have titles. And many of those posts are memes that have words in them.

He was taking a stab at visual design / aesthetics being prioritized over content when building a product.

Words are used simply here to better communicate his point.

Nowhere did he say that words are more important than photos or video.

You say "No" to him, but yet you know he would not disagree with what you have said regarding photos and video. I find the whole response to his original post borderline strawman.

Nowhere did the author actually say that.

I would argue the evidence for focus on "words" is overwhelming, and that you're attempting to change the meaning of the piece.

The author did say this however:

"At it's heart, web design should be about words. Words don't come after the design is done. Words are the beginning, the core, the focus."

The title is "Words"

The author uses "writing" and "reading" and "words" over and over again.

The word "content" isn't used once, nor anything like it (except as a negative, against content management systems).

And the author says:

"If you're a web designer, or a client who is working with one, I'd like to challenge you to think about words first. Instead of starting with a style guide or a Photoshop mockup, start with words on a page."

Again, the author is intentionally guiding toward a very precise thing: words. There is no confusion on the focus, it's blatant.

Right. The majority of images on webpages are the design elements, not the photos. His point would still stand if he included a relevant image inline. Focus goes to the content, not the chrome.

A good example - http://stallman.org/

Horrendous on the eye but gives everything you need. And more.

I think that if you replace "words" with "story" then he hit the nail on the head.

The web is about a story (not just words) and those can be told through a variety of mediums, but in this case, the author used words.

Your post reminds me of http://www.ftrain.com/wwic.html.

Show me a good, informative webpage without words.

I agree. I think the old way was about content now it's about a apps. One of those, if good, solves a problem.

Exactly.. without the graphics, audio, plugins, etc... it may as well be gopher, which most browsers don't even support anymore.

I don't get why the above post is so inflamatory to some... I used to use gopher, don't recall the ui all that well, but it was useful. Most browsers early on supported gopher as a protocol, and most do not today. And without the support of richer media (from images, to audio, scripting, even plugins) the web would not have been nearly as successful.

Agreed. If you want static text pages, go back to the 90s. We didn't feel very connected then, and there wasn't a big sense of community. It was mostly people publishing static pages for others to consume. Not much fun. Things have changed for a reason, and the internet has become a much more powerful place.

This article is absolutely pointless. Great, you didn't use a CMS, your content is now harder to manage. Users can't register or comment, so where do they discuss your article? Wait, they have to go to another site to do that, one that isn't completely dated.

This was a bit of an obsession of mine ever since I saw Coding Horror a few years ago. I had seen a lot of extremely minimalist designs, but they were all doing things to accomplish that minimalism. As a developer though, you have a nagging feeling in the back of your brain saying "that's a lie, it's not simple, it's very complicated, it just looks simple." Coding Horror is complex, but it was plain enough for me to imagine it with even less structure.

I decided to stick as close to plain HTML for my personal website as possible. The problem is, plain HTML is ugly. So instead, I tried to imagine what the default style of HTML would be if it was created today.

I've am still far from that ideal, but it is a work in progress: http://rkuykendall.com/

Your site looks great. I especially like your logo.

The logo is nice, but the aliased angled lines really kill it for me. I think a vector (or better anti-aliased) logo would really make a positive difference in the feel of the site.

The antialiasing on the logo was awful, and I've uploaded a new version. I think a vector version would be great. I am already serving up an @2x version which looks great.

I vectorized it for you: http://pastehtml.com/view/d6liy8ipy.html

All browsers have SVG support now, webpages need only consist of text, vector graphics and photographs.

it looks as if the colour is just a background-color attribute in CSS, which changes on mouseover. If so, it should be white->transparent anti-aliased in the RK image, which is looks like it's trying to be.

Vector, e.g. drawn with Canvas, would be better (e.g. with a fallback to a single colour png.)

It was trying to be, and failing. I am currently serving up a ( antialiasing fixed ) png, and @2x png for hdpi displays. Do you think a canvas logo would provide any additional benefit above these two?

If there was an error in the original anti aliasing that's now fixed, that should be fine - I was speculating how to get a vector version up and running (I am sure there are vector image formats, but all the web-friendly ones I've found turn out to be bitmap underneath.)

(just checked, after refreshing my cache a few times, and it looks much better now, nice one.)

What do you mean by "bitmap unterneath"?

uhh, I have been wrong about many things when it comes to web images - this is going back a long way. First, I always used JPG. Then, I discovered PNG, which I thought was vector (because I created Bezier curves in my editor, and when I exported them the quality was retained exactly). I can't remember what other formats I've tried in between (EPS and PS were involved somewhere, but they're not web friendly). I've never succeeded in creating a vector image that I can use on the web with a normal editor - although I have heard of SVG before. Maybe I'll look up how to export to it next time I need to solve this problem.

Why would you draw the vector manually with canvas when the browser can do it (SVG)?

I suggested canvas because that's how I've generated programmatic images from my own "vector" formats (scientific datafiles,) and rendered them at arbitrary zoom.

Generating the relevant shapes and lines for a simple logo would be easy this way, although probably would require a larger download vs. a vector image, since the canvas method would need to load extra javascript to duplicate the SVG support built in to the browser.

I'm not saying it is necessarily a better idea, just that I've had more success with canvas drawing methods than trying to actually make a vector image.

Making a vector image is piece of cake. The SVG syntax is a bit concise, but that's about it. See http://blogs.sitepointstatic.com/examples/tech/svg-curves/cu... . Regarding arbitrary zoom, it's just a matter of changing the viewbox: http://www.justinmccandless.com/blog/Making+Sense+of+SVG+vie... (it's no madness though) And you don't even have to redraw (that's the benefit of SVG, it has a scene graph unlike canvas). Remember, a vector image doesn't need to have any inherit width or height! In fact, if you don't specify it, the browser will resize the image so that the viewbox fills the browser viewport!

Another advantage is that you can handle generated images as images, not as code, without converting them to raster bitmaps (which I guess you'd be forced to when when dumping the canvas).

Thank you! I spent years without any kind of visual identification, but one day this summer I was playing around in photoshop and came up with that. It works quite well with a number of letter combinations and colors. If anyone reading this has trouble with personal designs, playing in an image editor whenever you have a stray thought will produce many terrible designs not worth saving, but pays off in the end.

Sadly you chose a font that only works when it's gauss-blurred.


I chose Source Sans Pro, probably the most fully developed, supported and profesional open typeface available.

What OS/Browser are you using that destroys type like that?

The OS/Browser don't matter. I disabled font smoothing and anyone who has that on on any software/hardware combination will be subjected to output like in my screenshot.

To explain this a tiny bit further: Fully developed and professional fonts will have had attention put into them to ensure that they will look perfect in pixel-exact representation at any font size. The font you chose had that kind of work only put in for up to font size 8 or thereabouts and thus looks terrible.

IMO: The logo mouseover is the one giant blaring loud thing on a page that's generally softer, more muted. I'd suggest you make the transition a bit more subtle, from dark grey to grayish blue rather than solid black to solid blue.

Thank you for the feedback. I plan on doing some more work on the site soon, and I will keep that in mind.

FYI - You didn't spell starting correctly in the sentence about your graduate studies.

That's what I get for trying to make a quick edit after my acceptance. Thanks.

Centered text, large white borders, and larger than standard font size are all design. Minimal design, but it does feel very different if you look at it completely unstyled:


I feel like it subverts the message a bit that he still felt the need to style things.

This is the full stylesheet:

  <style type="text/css">
  body { font-size:18px; }
  .wrapper { max-width: 600px; margin:0 auto; }
I feel like you're being a little unfair here.

That's quite beautiful for such a small stylesheet.

I once almost started a web site called SimpleStyleSheets that had that goal -- maximum of 4 lines of CSS, no ugly compression, maximum of 120 characters per line. And see how good you can make a page of simple semantic content look.

I should do it now -- it'd be a good exercise in this day of complex designs and bloated styles.

Beautiful? It's a tragicomedy! Even the most minimal stylesheet shows how unintuitive CSS is: To center content, your don't use the word "center"!

It could even be a little smaller if the max-width and margin were placed on the body tag. You don't even need the wrapper!

This wouldn't work with background colors though, would it? I imagine the document background color would be undefined outside the body then. Do you suggest putting the background color on the html element? I never saw that.

Still needed style. The good point is, we don't need to be extremist.

Maybe you are right, but maybe we can be extremist and it will still work:


ASCII text, copied from Firefox, added links and emphasis manually and formatted with par in vim.

Partially I was reacting to the words saying: "There's not much code on this page at all, just simple markup for paragraphs, hierarchy, and emphasis."

But the question is, should we see that as structure or design?

Let's kick it up a notch:

What we call unstyled really means that only the browsers default CSS is applied. To me that feels like the design of 1995, NCSA Mosaic, early Netscape and all that.

To get rid of that I applied Eric Meyer’s Reset CSS 2.0, it looks like this:


Although 1995 would have had a grey background.

Almost forgot about that. Ah... those memories!

I like your version even better. ;)

The point of the post wasn't: "un-style everything!"

The point was that simple, unadorned words on a page can be powerful and effective.

It's the difference between serving a steak with just a little salt to help complement the steak-ness, vs serving a barbecue sauce dumpling held together with cow protein. It's about whether the added style exists in order to augment the central content, or whether it's trying to be the main attraction in its own right.

The OP doesn't argue against styling. It argues against design cruft. The point is not to forgo design, but to focus on design that enhances content.

Words are the substance, style is the presentation. The focus should be on the content with style in a supporting role. You can't convey content without a style, so at the very least you should ensure that your styling doesn't detract from the message you're imparting. The first bite is with the eyes, as they say.

I couldn't agree with you more. It illustrates particularly well the fact that there is a line in the sand to be drawn and that it is ambiguous. His is just but one perspective on a site that serves up one type of content.

I don't think it really subverts the message. The default browser style is simply unusable given the widths most people have their browser set to. A minimal stylesheet to bring it back to usability is reasonable.

The point is that people have, at times, become a bit too obsessed with style, interactivity, pictures, colors, fonts, video, audio, etc. Sometimes, you just need to focus on the content. If a minimal stylesheet helps with that (just keeping a reasonable column width, and reasonable font size, to make the text readable), that doesn't really affect that message at all.


Oh wow he actually DID write that in a little city in British Columbia. I assumed he was pulling metadata and serving people different answers.

"I wrote these words, and you're reading them"... and Google knows you're reading them.

I find it a bit ironical that a manifesto for minimalism still carries Google Analytics code to track people.

It's true. There's a bit of CSS. There's also a tracking script.

I also needed a web server to host them, an FTP client to transfer them, plus the entire infrastructure of the web to make this work. ;)

So beneath the simplicity of "just words" is some complexity.

But all that stuff is invisible for most people visiting the page. The point is that simple words on a page can be powerful and effective.

How do I download your app so I can read the text better?

;) I'm working on it!

+1 Internet.

I don't see the connection, or the problem. Please explain.

I don't have a problem with it, but it was funny to me that when I viewed the source the first thing that I saw was some javascript.

So funny to me how people take the message so literally. Most people do not view the source of a webpage so a little script that doesn't affect the user experience is irrelevant to 99% of internet users. The message is about what you are presenting on the web...

I automatically checked the source after reading "There's not much code on this page at all, just simple markup for paragraphs, hierarchy, and emphasis." Just out of curiousity I guess. It's obviously not a big deal, it's just funny. It's not what I expected.

Part of the message of the text is about establishing connections — that someone reading the text the author wrote is "magical". The magic works both ways. The author should get to enjoy his analytics and "see" people reading his words. It's one of the great benefits of publishing on the web.

...and a DIV with a wrapper class. So really not "just" words.

"There's not much code on this page at all, just simple markup for paragraphs, hierarchy, and emphasis."

There's something magical about personal websites. Something that a Facebook profile page or a Twitter stream or a Tumblr will never reach. I really wish more people would go back to the earlier roots of the internet, and share what's on their mind in a more personal and genuine way.


"Art of Travel"[1] is one of those websites. When I read that website I was impressed how lovingly it was put together. The overall experience was so good that I didn't even think about evaluating its design.


I agree completely. Learning HTML was such an empowering feeling for me. You can make something fancy or minimal, whatever you feel like. We are all unique and personal websites show that much better than a social media website ever can IMHO

I miss MySpace pages too.

Reminded me of This Is the Title of This Story, Which Is Also Found Several Times in the Story Itself mixed with Web Design is 95% Typography



This is a sentence. That fragments. Useful.

One of my favorite of Hofstadter's essays. And thanks for your links.

I agree with him, and this is why I find Reddit, HN and some blogs I get to read around here (M. Gemmell's, M. Arment's, J. Gruber, PG) so good: they focus on the text, not on the fancy (and this nags me again to clear all cruft from my blog, but... some other day.)

PS: I kind of missed "all craftwardship is of the highest quality" in the page, though (or text that menaced with spikes of http or something.) I guess I'm too geeky today

Sure, simple text is fine for a blog. But if I'm selling something like a piece of art it's all about the images: large, high-quality images from multiple angles. Pinterest wouldn't be called minimalist, but it does a damn fine job of accomplishing its goals.

Design should help accomplish business (or personal) goals. We run into trouble when we adopt some sort of "minimalism, always, ever, for everything" dogma.

I think the author clearly states: "Think about the words first, than the rest".

I didn't read it as a call to minimalism, more so as "It does not matter how flashy your site is, if in the end you have really nothing to say."

It's hard to disagree, honestly (it's also plain common sense, nothing groundbreaking, although the minimalist presentation really helps driving home the point).

Completely agree

Just s/words/content, if you like.

If your content is pictures, the same philosophy applies: ditch any ornamentation that gets in the way, just post your damn pictures. If you're selling the pictures? Get the pictures out there and get the "buy" controls out there.

It's still minimalism, and all it's saying is: "if your goal is to show pictures, that fancy layout that pushes all your content into a slick grid of thumbnails, just so people can see some slick fly-out animation when they're clicked upon, is probably just wasting everyone's time."


Relying on words conflicts with the popular concept "show, don't tell". So I think the author's point only applies to a subset of web pages. From my perspective, I want to see screenshots of your app (or preferably try it with a non-signup demo) - I don't want to read a bleeding heart story about your motivation for it. Give me a tagline at best for that.

App web pages need to be designed like HyperCards or PowerPoint presentations - where you aim to lower the number of points on the screen to retain attention.

> We run into trouble when we adopt some sort of "minimalism, always, ever, for everything" dogma.

Evidence? Examples?

Is it not self-obvious that if every page attempting to serve up content similar to that in the OP (pictureless editorial), the web would be bleak and characterless?

Not debating the effectiveness of that kind of design, but rather pointing out the obviousness that this sort of anti-"always, ever" statement is easily validated.

Most of the web is awash in ugly junk. Whenever I go to a machine that doesn't have adblock, Ghostery etc., I am increasingly horrified. It's like a highway with so many billboards that you can't see the direction signs, never mind enjoy the view. That or some sort of infinite supermarket magazine rack.

I HATE the way web design has evolved. I want to make the design choices at my end, not have them imposed on me or consuming a ridiculous amount of bandwidth. Most of the design on the web is no good.

"Only on TV and radio. And in magazines and movies and at ball games, on buses and milk cartons and T-shirts and bananas and written on the sky. But not in dreams."

I imagine a lot of users would love to have a web browser alternative or software extension that gave users complete creative control in an accessible way- including advertisements, colors, typography and margins.

I think such a software would have to be free, but I also believe it would change the way web design and advertising works if it caught on.

Having webpages that are not bleak and characterless is not a resignation to having webpages that are like highways with unending billboards.

I don't find text to be bleak or characterless; if it needs design that badly, then it's not well-written. And judging by the audience for things like ReadItLater and so on, I'm not alone. I just want the option to read text the way I want, not how someone thinks I ought to be see it, which is more often about getting me to look that providing me with a comfortable reading experience.

I didn't say text is bleak and characterless. I said that having no design variation is bleak and characterless.

Then style the web as you fit on your own browser, but don't try to impose your design sensibilities on mine. That's why I keep saying 'client side'. I prefer not to download and process most of the graphic cruft that I see, and vastly prefer reading articles in an RSS reader to visitng pages that remind me of being a shopping mall.

You are already able to do this; simply disable stylesheets or as you said use a reader app. We've already discussed using technology like adblocker.

Your issue therefore is not with the design but rather with the browser. There is nothing preventing you from using a text-based browser or even writing your own.

I do not know if you are a developer, designer, or something else, but you must realize that minimalism is itself a "design sensibility" as you term it.

I appreciate where you're coming from and I'm sorry if my earlier comments came off as snippy. But I do think the influence of design on the web has been too heavily tilted towards the publisher end and not enough towards enabling the consumer.

I have the skills and experience to retool my browsing experience for my more minimalist aesthetic, but most consumers experience the web as something more like an animated version of a supermarket checkout magazine. Since getting a smartphone a few years back, I've been especially struck by how much bandwidth is consumed on transmitting cruft on things like news stories, where the actual text is only a few kilobytes and relevant photographs take up no more than a few hundred kb; but after you pile on all the trackers, divs, ads, etc. etc. the page ends up being many megabytes and can easily take 30 seconds over a typical 3g connection.

To me that's enormously consumer-hostile to very little real publisher benefit; the more shit I have to download to see a page the less likely I am to consume news from that source and the more likely I am to reinforce my ad blockers etc. (whereas I will make exceptions for site that make restrained use of such things so they can get a little extra ad revenue or tracking data value in return for the service they provide).

I agree that ads can be garish. I just see this as a separate problem from design itself.

Recently XKCD made me think about how minimalism fits into design. For example, entirely text-driven:


... and more visually oriented:


Obviously there are many ways to communicate ideas, and each in various contexts can be quite effective.

If every page was similar to this one, the web would be plainer and simpler. It seems that you seem to think this is a bad thing by characterizing it as "bleak and characterless." I think it would just be a whole lot less stupid :D.

Pretty much everything I read on my Kindle looks the same in terms of font, styling, etc. Far from being "bleak and characterless", it gets out the way and lets me read the story, which is what I'm there for.

Unix-style "bunch of small processes communicating through pipes" model when applied in inappropriate domains.

Use only the text editor ed for your next project. Oh, and no version control. Don't tell me that's impossible, either, because we both know it isn't.

That's a stupid analogy dude.

Why? It's minimalism, isn't it?

I agree.

"If additional styling, image, media, or script gives the audience something more - then add it."

But that's what minimalism is. You can say the same thing in reverse.. "If it doesn't give the audience more understanding - then leave it out."


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact