I think everybody who's spent any time in the industry will agree with the feeling to some level. That being said the "harfmul stuff" table is a prime example in flamewar-bait, I seriously hope nobody will take it at face value. CVS better than SVN? Anything from plan9 instead of stuff people actually use? Good luck getting 9p to work reliably faster than NFS... Oh and you should use ed instead of vim, tcc instead of gcc and awk intead of ruby.
If this table is not meant as pure bait then I think it's actually harmful on its own, an experienced dev will be able to pick the right tool for the right job on their own. On the other hand a beginner will be tempted to cargo cult it to seem hardcore without understanding the tradeoffs.
Sometimes it's simpler to reuse a complex solution than write your own simple code from scratch. It's easier and simpler to use C++'s std::map than implement a red-black tree from scratch in C.
>Good luck getting 9p to work reliably faster than NFS
Fast != good. A slower system which works more reliably and elegantly is often (usually) a better choice.
>awk intead of ruby
I agree with this, though it's expressed poorly. For use-cases within the intersection of awk and ruby, use awk. For use-cases without, you probably shouldn't use ruby anyway.
>It's easier and simpler to use C++'s std::map than implement a red-black tree from scratch in C.
Easier, yes, simpler, no. When you use C++ you get all of the baggage that comes with it, increasing your complexity hugely.
> That being said the "harfmul stuff" table is a prime example in flamewar-bait
Agreed. Especially with the GPL and LGPL, etc being "harmful". That seems like they're trying to evoke an emotional response.
Sure, if there's different software I should use instead of $popular_X , I'm all ears. But be aware, there's context switching costs from going to one to another. And some software, even being worse, has better support by the community.
Im going to flag this article, for your complaints. It's not constructive, and spawns flame-y yelling.
>Good luck getting 9p to work reliably faster than NFS...
Could you elaborate on that? From what I've heard (in my echo chamber), NFS is "complex, slow, and unreliable" while 9P2000 is "simple, fast, and reliable".
Technically there is RFC4180[1], but yeah, CSV has a lot of flavours. I remember a story about how Excel couldn't correctly open a CSV file because it had a header that had "ID" as the first field name, and Excel's file format sniffing algorithm though that it's a different format.
Technically RFC4180 isn't actually a standard but perhaps the world would be better if people could agree to treat it as such. However those living in countries that use the comma as a decimal separator may disagree.
> This memo provides information for the Internet community. It does not specify an Internet standard of any kind.
As someone from one of those countries I've never understood this criticism of CSV. JSON doesn't let you choose your own separator either but no one levels this complaint at it. Defensively quoting values that may contain commas is easy, and nearly always a negligible cost. This really seems like a total non-issue to me and I've never understood how it has gained so much mindshare among CSV critics.
The problem is Excel's horrible CSV implementation, and opening the file in Excel is one of the main applications of CSV. The German version of Excel expects semicolon separated CSV, where comma serves are decimal point and the date format must also be German. I found no way to create a CSV file that opens in Excel, regardless of locale. (And there are a million other problems in Excel's CSV implementation)
Perhaps because there are a multitude of, sometimes ill defined or ad hoc, specifications that overlap one another in both subtle and not so subtle ways.
So you can't assume CSV files produced by one program can always be read by another.
Uriel, the creator of this page, has committed suicide in 2012[1]. As far as I know, the website hasn't seen that much support since then. We could only guess, what he would've added to the list today.
My favourite page is the one with programming quotes[2]. I don't know, who that Al Viro is, but he is quite a savage.
I'm known to my coworkers to complain about things I dislike and that I dislike almost everything.
This page lists things even I like.
It doesn't even go on to explain why the things are better or worse. Well, it sort've does, if you look for the bad things on the left side. But then he goes into FUD from decades ago that's particularly not relevant to today (in C++ specifically, probably also true for other bad things).
I'm all for ranting about terrible software but at least make sure it's accurate and versioned/dated.
So this is a list of "harmful things" and "less harmful alternatives", with no explanation given as to how each tech has been classified. To take a random example, why is man less harmful than info?
Fun little story, I used to be sympathetic to the all software sucks movement and attitude. Then I tried to port `st` (simple terminal) to Mac OS X by separating its core from its X11-dependent code and pulling the globals into a single struct, so that I could wrap it with NSView. When I tried to ask the `st` community if they wanted my changes[1], they replied that these changes were unnecessary, that coupling the core with X11 was perfectly fine, and especially that there's no reason to port it to such an awful non-free OS as Apple's. I now prefer to take a constructive and positive approach to software development, instead of just having a negative or cynical attitude.
I can't believe how rude the comments in this thread were.
I generally agree that we have too much complexity in software, but I also feel that often people take the "minimal" software notion to an extreme level of elitism, excluding and attacking people for reasons not related to the original goal to make better software.
It progresses toward constructiveness near the end of the thread though, FWIW, so it's worth reading through if you want sort-of-closure I guess.
My only disagreement is with the one person who said that the other people in the group were strongly-opinionated and that it was "healthy". No. Some of the messages in this thread are deeply insulting personal attacks and simply inappropriate to ever submit anywhere.
That you managed to stay focused is impressive. I fear I would have spent _way_ too long on, shall we say, an "appropriate" reply. Actually I might've just up and left, heh.
I appreciate the TIL regarding a community to call out and avoid. Yes, people like the ones found in this thread are everywhere, but to paraphrase something I said on a similar subject recently (https://news.ycombinator.com/item?id=17730570, see parent comments for context), I have no inclination to field the signal/noise ratio to engage communities where such people are majority vocal. With such people, it seems you have two options - argue against their views and suffer the continuous backlash; or implicitly agree by not arguing, and risk the potential expectation to behave contrary to your personal standards as a result. Alternatively, one simply avoids them, or calls them out.
HN's moderation does a reasonably good job of eliminating such contention. As time goes by I continue to find new ways to appreciate the carefully-curated status quo here. It's certainly not perfect (particularly the automated post-killing system, which kills a small but noteworthy number of good posts) but at least I never have to drown in distraction and meta basically ever, and I have showdead on so I can read the dead comments (and, most importantly, read with the expectation that they might be inflammatory so there's less surprise), so it's pretty good.
> I fear I would have spent _way_ too long on, shall we say, an "appropriate" reply.
When someone's only interest is in being right and explaining how right they are, it's not worth talking to them when you disagree. You can listen of course, but that doesn't need a conversation.
> I appreciate the TIL regarding a community to call out and avoid.
This was 3-4 years ago, I'm sure the community has changed at least a little since then, like any online community does. It also may be completely different today.
> With such people, it seems you have two options - argue against their views and suffer the continuous backlash; or implicitly agree by not arguing, and risk the potential expectation to behave contrary to your personal standards as a result.
Silence is not implied assent. Silence is the right solution to most situations. Having been on the arrogant end of those kinds of "debates" for far too much of my life, I've learned that silence is effective because it does not give the arguer any satisfaction in their prior arguments nor the opportunity to continue showing off their arrogance with further arguments.
> HN's moderation does a reasonably good job
Having been in many, many online communities for the past 20 years, HN is in fact a minor miracle. It is a bit of an echo chamber, but you can't really escape that while still having a "peaceful" community. cake.eat(); cake.have(); throws "Error: cake reference has become invalid"
> This was 3-4 years ago, I'm sure the community has changed at least a little since then, like any online community does. It also may be completely different today.
Ah. Duly noted, thanks.
> Silence is not implied assent. Silence is the right solution to most situations. Having been on the arrogant end of those kinds of "debates" for far too much of my life, I've learned that silence is effective because it does not give the arguer any satisfaction in their prior arguments nor the opportunity to continue showing off their arrogance with further arguments.
> When someone's only interest is in being right and explaining how right they are, it's not worth talking to them when you disagree. You can listen of course, but that doesn't need a conversation.
Thanks for this, particularly the first sentence. I'm still learning to mentally model behavior so I can understand the thinking behind a given form of expression, and in turn how I should respond appropriately to it. Now to just learn how to actually quieten down in these settings... :)
> Having been in many, many online communities for the past 20 years, HN is in fact a minor miracle. It is a bit of an echo chamber, but you can't really escape that while still having a "peaceful" community.
For better or worse, I'm really glad I was having slow internet a couple years ago and went "alright what's the URL of that really barebones website again, that'll probably be usably fast" :D
The SnR here is probably the reason I like the place so much; reddit is full of _so much fluff_. Hm, maybe the reason people say they come here for the comments is not so much because the comments will have exactly the thing they want to read but because they're not full of distractions...
> cake.eat(); cake.have(); throws "Error: cake reference has become invalid"
I use dwm, st and slock myself so I'm definitely grateful for the suckless people but it's true that their arrogant attitude is often silly, unwarranted and counterproductive. I also find that they often mistake terseness for elegance and simplicity, while their code is decent it's far from the best C code I've ever encountered (apparently tests and comments are not part of the philosophy). Their fetishization of C is also ridiculous, they actually advise using C instead of a shell script to set the dvm status bar because it's more minimalistic that way. Since you'll probably want to read a bunch of /proc and /sys entries and/or call third party programs I don't really see why anybody would think C is more appropriate than /bin/sh.
But hey, I'll still take that over "what's wrong with Electron apps anyway?".
Just do what I do: fork their code and never even consider contributing anything back since you'll probably get shut down anyway.
Listing FreeBSD, D, Java, C++, and SQL databases as "harmful stuff", and suggesting "hierarchical file systems" and "textual interfaces" as not harmful. What the actual hell?
I’m sure we’ve all seen the various famous examples of this kind of opinion and attitude floating around the Internet over time, but I’m not sure what value they really have. They’re mostly unhelpful aphorisms that don’t really offer any context or argument.
The “less harmful alternatives” are simpler only in a surface-level way. Go is “simpler” than Python in that the language has less features, but that leads to more complex code since there is less possibility to abstract things away. It is also a very hard language to write reliable code in, and reliability seems to be the stated goal of this whole thing. Python is hard to write reliable code in too, but Go is not a good alternative in this aspect.
Also, many of these alternatives make no compromise for practicality. GCC sucks, we can all agree, but without it our code would be slow and buggy (TCC doesn’t support ASan or other such debugging tools AFAIK).
> At the moment a detailed rationale is not provided for most of this, so figuring out why some things are considered more or less harmful than others is left as an exercise for the reader. Here is a hint: complexity is the bane of all software, simplicity is the most important quality. See also: Worse is Better by Richard Gabriel.
This is, well, just another way for people to turn their opinions into objective fact. The lumping of C++, D, Python and Ruby all together betrays either a lack of sincerity, or over-opinionation to the point of myopia.
The page links to Zed's rant about how Rails is a ghetto, while he makes some great points about the economics of that ecosystem, I'm left with no salient reasons why I shouldn't choose Rails for a project. Sure, if I want to do what he tried to do in that space, I have no reason to not expect I'll have the same kind of experience he did. The Rails community might be a ghetto, but my project doesn't have to be.
Read between the lines and you see Zed does have respect for Ruby and the Zen of the language. To see it linked on cat-v, to my mind looking for real reasons why people should and shouldn't do things, well, the essay clashes pretty hard with the seeming ideology that "complexity is the bane of all software." Zed was perfectly fine with Rails and all its complexity, he just hated the community and ecosystem. Cat-V is linking to his essay not because they actually agree with him, but because they are enamored with his style.
The whole site reads like a bunch of neckbeards blowing off steam.
Not sure who prepared this list but obviously this is extremely subjective.
C is less harmful alternative to Java? Are you kiddin' me? Have you ever tried creating any non-trivial web service with ANSI C?
For almost every example on the list I could bring examples showing situation where exactly opposite is true.
The old rule of thumb is, any definitive and unsubstantiated statements about superiority of any technology over any other technology should not be trusted.
Years ago, I was complaining to a co-worker about a terrible codebase. He looked at me and said, "ok, when was the last time you looked at a program and thought, 'wow, this is really good code'?" That simple statement changed the way I view software quality forever.
"Less harmful alternatives" seems in many cases to be "Exotic technologies hardly used by anyone, simply unusable in any serious environment". In some cases, comically wrong: C as a better alternative to C++ ? Did I totally miss the obvious satirical nature of the page ?
It's still used today because (1) of historical code and (2) the fact many people are fluent with it (including me). But this is a terrible languages in most cases.
C is relatively simple, C++ is so complex practically no-one understands all of it. Many C++ features are "OOP kool-aid" and do more harm than good (if people would just realize you can pass around pointers to C struct's, which is all the "object orientation" necessary for 99% of the software in the world)
You can do in C++ everything you can do in C. You don't need to use C++-specific stuff.
Also, using your logic, C constructs are nothing more than just assembly with more overhead. Why not just use assembly instead? Answering this question will probably answer your question about OOP stuff.
By that logic, we should all use "omniglot", one language in which all the top 100 languages are first-class supported. What could possibly go wrong :D
If this table is not meant as pure bait then I think it's actually harmful on its own, an experienced dev will be able to pick the right tool for the right job on their own. On the other hand a beginner will be tempted to cargo cult it to seem hardcore without understanding the tradeoffs.
Sometimes it's simpler to reuse a complex solution than write your own simple code from scratch. It's easier and simpler to use C++'s std::map than implement a red-black tree from scratch in C.