Hacker News new | past | comments | ask | show | jobs | submit login
The new word processor wars: A fresh crop of productivity apps (geekwire.com)
131 points by prostoalex 3 months ago | hide | past | web | favorite | 104 comments

Just a quibble but I see this a lot - if you're talking about strategy it's "bottom-up", not "bottoms-up".

A bottom-up strategy is something like marketing your tool to users who then bring it into their company for wider adoption.

A bottoms-up strategy is drinking until the problem goes away.

you shouldn't discount the power of the second one

As a recovering alcoholic, I can attest that drinking does not resolve life's problems.

I, for one, am glad you are recovering.

Hah! I've def made this mistake

I have the same feeling about this, and I've noticed that folks from the UK are more likely to use "bottoms-up" to mean both. I rarely hear it from US-born speakers. Has anyone else noticed this pattern?

I'm British,

Never heard "bottoms-up" to mean the same as "bottom up"

Bottoms up is when you finish your drink.

Bottom up is sitting on a high surface (put your bottom up, usually said to a child), or lifting something from the inverse of where it's normally handled from or, like the parent said: a set of business strategies.

Surprised to get downvoted for asking a question to see if others have found the same thing. I'm not passing judgment — I have a background in linguistics and see grammar/usage as descriptive, not prescriptive. I just wondered if usage was different in the UK. Sorry if this was misconstrued as passing judgment, or something else that apparently warrants downvotes.

Not a Brit myself, but I've noticed they tend to put an s on the end of a lot of things. e.g. "maths"

"Maths" makes sense though, as short for "mathematics". However, "bottom-up" refers to "from the bottom up", I think, which is singular.

At this point I look at Word and Excel the same way I look at bash, gcc, grep, sed, awk, et cetera: there is so much iniertia that even when different and arguably better tools come along, it still never truly goes away.

A lot of people still use emacs, vim, BBEdit and so on, but a lot of people also use IntelliJ, VSCode and Atom or whatever is in vogue this year. Vim and emacs will never truly go away no matter how many CPU architectures or operating systems come along this century, and neither will Word. At some point software is just infrastructure. We can build better bridges than we could a century ago, but there isn’t an incentive to replace a bridge until it falls down, and it isn’t like Word prevented Markdown and TeX and HTML from finding their niches.

Word, yes. Word is a turd. Word needs to die.

Excel, however, is magic. Forget the nasty ribbon interface and the horrid MS window chrome, I'm just talking about what happens inside those cells.

Say what you will about Microsoft as a company, about the fact that Excel is possibly over-used and that users should probably get a database. I don't care. Excel, at its core, is a magical product. Nothing else gets close to its functionality/friction curve (assuming you already know how to Excel, that is).

Not even inside the cells. Excel column / rows filters. Hard to do better than those. I wish a lot more apps with data implemented something half as good.

Is this truly attributable to what Excel is actually capable of (excluding those cases like people who manage to accomplish things like 3D rendering with excel) or that Excel just has better market saturation and is, by default, the only spreadsheet tool people in the Enterprise are given to work with and we just normalized ourselves to what it does, but maybe not so many other spreadsheet file formats?

I also think Excel is great; but not because it's Excel or heavens forbid, because it's a Microsoft product, but because it's a spreadsheet.

The ability to throw data in a grid and massage it with operations the average human can understand - that's invaluable.

Certainly, Excel is powerful and feature-filled. But 95% or more of what I do in Excel I can also do in Google Sheets or Libre Calc or whatever.

This. All of what I'd ever want to do to a spreadsheet Google Sheets can do, and whatever the 'power' Excel user thinks is a good idea to use Excel for I can do far more easily in R.

You can. G'donya. If you print a report on the result - you're done. Bottoms up, mate. If you are collaborating - the issue isn't tools that you and I might know that may or may not be better than Excel (like R), it's that we've got to share our work with people who may or may not know those tools.

I believe (personally) that Excel almost singularly is a tool that both the adept and novice can use together. And, you know, if you want to spin your twirly six shooter, you can always do a monte carlo simulation in C, print out the results as CSV and import it into Excel for the (I'm sure you would call them) more benighted amongst your co-workers.

Except times are changing, and maybe we won't see it until baby boomers are all retired, but a lot of younger people are far more adept at using technology and coding is becoming far more popular. In university, we had to learn Python and SQL in a CS for Business introduction course, R for stats, and while we did learn some Excel, for anything related to stats, econ or even finance we used other tools primarily. Just like MS woke up one day to the fact it no longer has a strangle hold on computing, the same will happen to Excel, but maybe in 10 years. There are better tools and the people who cling to Excel are on the way out.

I said the same to myself as I looked over my shoulder at the millennials in ~2000. I was sure that the couple of glorious working years I’d had so far – glorious because of the rate I was getting compared to the work I had to do – would soon be over.

It never happened. The millennials, as a whole, don’t sit in front of Python any more than my generation do (I’m 42). And if anything, the number that do is getting lower. At least I had a ZX Spectrum 48+ to hack around with when I was a kid. I had to code to load a game, even if it was just `10 LOAD`! Kids now just hit a button on their iPad.

> we've got to share our work with people who may or may not know those tools

We can acknowledge the power of network effects while at the same time critiquing design. They both have utility, but they are entirely separate.

> Word, yes. Word is a turd. Word needs to die.

It's not clear you'd find a lot of lawyers (i.e., a class of power small-document-scale Word users) that would agree with this sentiment.

I am an attorney and I use Word exclusively and intensively. And I hate it. All of my colleagues hate it too.

What would your preference be?

Mh, I have sometimes to work with "tabular datas" that seems to fit well with spreadsheets... Before I discover org-mode and Emacs I normally use *sv sources/exports and CLI tools simply because I never being able to be productive or in comfort with ANY spreadsheet I tried. Now I'm in org with tables and babel I never look back.

IMO spreadsheets are a poor and ineffective answer to the miss of a serious desktop database tool with a nice DSL for end users.

My two weapons of choice:

Excel for structured data, Emacs for unstructured data.

Why the hell a spreadsheet? Why not use any scripting language you want with org-table/org-babel or with some ORM/driver to access datas?

Because it's convenient, and I can share.

Not to mention the fact that Excel isn't "just a spreadsheet".

I could go on, but most of the other sibling comments here have expounded enough upon her virtues.

EDIT LOL ORM/driver - because life's too short that's why

Mh, for me just because life is too short I never really learned spreadsheet software (and I have seen many disaster made thanks to them, including important institutional and scientific ones) and I'm happy without them...

Try to look for some Emacs demos on YT&c, perhaps you'll discover a different world :-)

> Try to look for some Emacs demos on YT&c, perhaps you'll discover a different world :-)


I'm already a fairly proficient Emacs user.

I also use vi.

I find this kind of zealotry tiresome.

Mh, so you find more comfortable work in a spreadsheet than an org-mode files with tables and optional org-babel block in your preferred language to manipulate data as native data structure if org table formulas are not enough nor comfortable?

Personally I think there is no comparison, spreadsheets are normally uncomfortable UI to work with datas that even miller+bc&pipes is far superior... Org bring computation to another level...

> you find more comfortable work in a spreadsheet


Is that really a solid argument? Haven't you also seen numerous "disasters made" with other platforms / tools / programming languages "including in important institutional and scientific ones"? (no snark intended)

Well, yes but far less... I see Ariane 5 disaster due to an overflow error, but never seen the International monetary found that publish a false study due to an excel calculus bug nor scientific studies with false result due to excel bugs...

Spreadsheets IME are environment that really help human error and disguise them well enough...

About solidity I can't answer having not enough data, I can only say that I still have to see anything more efficient than Emacs in the world and I see software since around twenty years...

The IMF error was a problem with the way that the humans had weighted their numbers in Excel. It was a human error. Excel was the tool they used, but it was not the cause of the error.

Yes, but it's UI, not as "excel" but as "spreadsheet" certainly help a lot... Try to "debug" a not so small spreadsheet vs a script+tabular data and compare the chances of get erroneous results that does not appear immediately as errors...

Why don't you like Word?

I think it's still going to be the best for a while, if not only because it's massive. My CS professor says it's the biggest program ever written.

It can so so much, like check for passive voice and more subtle grammar things.

At the end of the day, the subscription model is off-putting, but "needs to die?" why?

> Why don't you like Word?

Here are my main reasons:

- It does not allow me to make selections where I point my mouse cursor.

- It breaks between versions of itself and intentionally breaks interoperability with nearly all third party software, especially when tables and formulas are involved.

- Formula and math editing and typesetting is abysmal.

- The quality of typesetting is abysmal. I know because I once had to convert an article with formulas written in LaTeX by hand into Word. The editor of the journal was very ashamed of the quality of the result in comparison to the original.

- Images and other boxes inserted into the text almost never work in the intended way. They slip between pages, get the wrong flow around them, cannot be selected, etc.

- It is practically impossible to import documents produced in other tools reliably into Word or reliably export Word documents to be processed with other tools (like e.g. LaTeX). Even pandoc cannot compile to Word files that work reliably enough. Some of the errors that occur are nearly unfixable, e.g. bizarre things strange characters in lines that cannot be erased, stretched and distorted characters.

- Horrible default auto styles.

- Ribbon interface and other user interface problems that make Word one of the most unintuitive pieces of software on earth (many people just don't realize this, because they have been using it for so long and so often)

- relatively expensive

> Formula and math editing

Word supports LaTeX for formulas now!

That aside, Word's formula editor is really powerful but it does have a learning curve. I'd say that once learned it is rather nice though.

FWIW Unicode Math, Word's format for doing math input, is a published spec.

Because it tries to do far too much and fails at the basics.

Keeping track of the margin/indent settings for bulleted lists across a 50+ page document? Shambles.

Trying to guess which header you want and insisting on auto-formatting the wrong one, time and time again? Infuriating.

Weird behaviour where it promotes the penultimate paragraph to a section header if you delete the last paragraph of a section?

Deeply confusing and hidden ‘Styles’ functionality that affects the visual layout of your document in ways that a beginner couldn’t possibly understand?

I wish I’d kept a blog of Word’s infuriating behaviour over the years (except at the time, all I want is for the thing to let me finish the document I’m being forced to work on so that I can stop using it as soon as possible).

And they just keep shovelling shit in to the product. STOP IT FOR THE LOVE OF GOD.

The worst part is that for 95% of the cases, Word is entirely unnecessary. I’ve seen – god help me, I’ve written – 100+ page corporate documents that could just have easily been put together in Wordpad using RTF. But no, it’s got to be in Word. The thing is a fucking curse.

Word is hard for detail-oriented people like programmers. I want to be able to control and tweak every single bit just the way I like it, and it frustrates me that it just seems to think it knows better a lot of the time, or that it doesn't let me undo its mistakes in any obvious way.

But when it comes out at the end, it usually looks fine. I try to remember that it's the paper-based outcome that Word is designed around, so ultimately for the intermediate digital artefacts "just fine" is good enough.

I'm not saying it's right, but most day-to-day users are probably fine with it.

1. Your CS professor does not sound very good at his job judging purely from the very tiny amount of information you have shared. I am sorry you have to deal with that.

2. I, at no point, stated or even implied that Word needs to die. I am largely indifferent towards Word. It is a piece of infrastructure that other people rely on, and as such, it will continue to have a place on the market. My point is that it was Infrastructure at all, not to criticize others for using it, or Microsoft for continuing to develop it, or the Office team for their work on it. This is true for many projects, and overall there seems to be a type of software which becomes infrastructure and will never be replaced even when software critics believe it should be replaced.

Well... Because is a complex, utter, inflexible software.

For instance: I need to write a letter, on my desktop I start skeletor (Emacs built-in template solution) with a single keypress, choose the wright template, start writing in pure text, hit C-c C-e and choose the appropriate export format, for instance pdf via LaTeX. In a fraction of time I made a document far superior of any word-processor can ever made, without wasting time in formatting. Same if I have to create slides. Same if I have to mix tabular data, literate stuff like write down a formula or a software procedure and verify it live from the code inside the document etc.

The level of ease, flexibility, simplicity, reusability, integration you can have on text-centric systems you can't even imaging to reach with modern/classic GUIs.

Another example from my personal workflow: sometime I have to put notes in agenda, like TODO: fix xyz etc. I can simply add relevant SCHEDULED or DEADLINE or whatever in the file I'm working on and see it properly in agenda. I can have documents spread on multiple files for many reason but being presented and exported as a single one or vice versa. I can link emails anywhere, export them inside the document etc.

No other IDE, PIM suite, office suite can do the same.

> hit C-c C-e

I have no idea what this even means so it’s already too complicated for me to follow!

Well. Basically if I have to write a letter I:

- hit a single key on my keyboard (pause to be precise)

- hit 'l' followed by few arrow down to choose between italian, french, swedish or english letter template than enter

- start type the letter (my sender address autocompleted, I only type firs letter than tab and yasnippet insert the rest correctly)

- hit a two key combination (control+c than control+e) or run org-export and quickly answer (with predefined defaults) where to save my file than create and open a nice pdf ready to print, optional scanned signature included.

Of course that work anywhere on my desktop, I do not have to open any specific application, choose templates, choose from them, fight to format the text properly, find in complex menu how to insert an image (my signature), go looking for it on disk etc. Of course my letter get properly saved/archived digitally, I can send via mail in other two keystrokes etc.

It's normal that sound complicated for you if you never tried it. I see many colleges that remain astonished seeing my demos inside Emacs and can't hardly figure out what happen on my desktop. It's a path to learn something new and effective, but learn an OS, a different system not a simple single-purpose application.

Well I'm sure you must have good reasons for setting something up like this and it must work for you, but to me it all sounds like mysterious incantations (why are you pressing the pause key to do something other than pause something?) and over-optimisation, rather than making things plain and simple.

How many letters do you write a day for it to need to be optimised into single key shortcuts?

My bottleneck is never formatting a letter - it's deciding what to write.

Also: what I actually meant was I didn't recognise C-X as meaning press the control key and then the X key. I've always seen that written as Ctrl + X. Why does Emacs need to do it differently to everyone else?

> why are you pressing the pause key to do something other than pause something?

because I do not use pause key for any purpose and single-key action are super-comfortable. For instance in my setup I open my mail dashboard (notmuch-emacs) with it's saved searches that I can open with a mouse click or a single key action again), tags etc simply hitting F6, shift+F6 open a mail compose buffer (window), F10 open firefox with a work profile, shift+F10 firefox with a personal profile, F7 open agenda, shift+F7 open agenda's main file etc.

It's far simpler, efficient and comfortable than click on some launchers that are always less than you need. Only I feel the pain that actual keyboards offer far less keys than in the past...

> How many letters do you write a day for it to need to be optimized into single key shortcuts?

I write them rarely, but I have no shortcut for letter but for ANY documents, the pause keys simply popup a menu that present me some personal choices (letter, reports, slides, projects of various kind and languages etc), I choose one from the list and find it pre-made as much as possible. Only for letter I use normally double-window envelope that to work properly demand a certain position, certain address format, certain fonts etc. If I made them by hand and rarely I certainly make mistake, have template them once I can write straight away painless. It does not cost me substantially anything, only around half an hour once in my life and pay back at any letter I write. Same for slides. It took few hours to made few personal template and after I can only write down contents. Same for thesis, reports, ...

Basically a step at a time I crate my digital environment thanks to Emacs and now I have anything at my fingertips at an unprecedented level and essentially at no cost since I add stuff when I have time, calm, without pressure nor frustration and I found it back when I need and perhaps I'm on hurry.

For instance I have regular invoices by mail (telephone, electricity, water supply bills) in the past I spend few time regularly to read them (I normally have them auto-paid by the bank) and archive by hand. Now thanks to notmuch hooks I have simple scripts that when a known bill arrive archive it accordingly and add a small note to my startup dashboard. So now when I have time I read the bill with a simple enter in the dashboard, when I quit the entry goes away from the dashboard and anything is done. If the correspondent payment does not show up in ledger (personal accounting) after some time (actually I'm not developed anything that guess when to expect the payment from an invoice) an alert popup in my dashboard and when I open it I found in a click/enter the "unmatched" invoice...

I have similar automation, all made in free time when I'm willing to build, for mail autorefiling (IMAPFilter), with Alt+F6 (my "official mail key") I go straight to unread mails or get a "no unread mail" message in the minibuffer echo area and mail are automatically refiled as much as possible in their proper dir. After a certain amount of time and depending of mail directory IMAPFilter automatically delete the message helping me keep my mails clean and easy to search. When an unread message need to be left around for some reason I may add a proper tag to snooze or categorize it in some temporary way, few tags can be added by single keypress on the message, for instance 'l' add "live" tags that automatically show up in general dashboard, and mail dashboard, O add "order" tags for online shopping done but waiting to receive the package, T add todo tag etc.

It's more complicate to explain it than build thanks to Emacs and again I'v build in baby step and more other will come, when I feel the need for it or I'm curious or simply I have free time and willingness to automate some things.

That's in synthesis an ecosystem tailored to my needs, thanks to flexible and easy to customize software, that keep me evolving everyday. All other environment I know are like a wrench: can be useful for doing few things but can't really be customized nor evolved.

I thought emacs is a text editor, it appears here you're using it as an OS? do you boot directly into emacs?

Something like that, Emacs is my windows manager and start at boot with autologin (I use LUKS crypto root so I do not need to type a password on personal systems). So yes, while I'm using GNU/Linux (NixOS) I can say to a certain extent that I boot directly into Emacs.

And it's not a new thing BTW, many years ago (early 2000s) there where already an Emacs WM (http://www.nongnu.org/xwem) and someone already use Emacs even as a login shell instead of bash/zsh (http://www.sxemacs.org)

You can consider Emacs as a text-centric user environment so the "editor" part is of course there :-)

Try looking on YT about "Notmuch: What email should be", "Introduction to org-ref", "literate DevOps", "Emacs reproducible research", "Conquering Your Finances with Emacs and Ledger" etc they are about few use of Emacs and often they are presented in Emacs itself, sometimes as a WM with the modern EXWM :-)

Also GuixSD (guix emacs package) and Arch (arch-packer) have system management tools inside Emacs, other generic tools exits like proced (top/htop inside Emacs), enwc (NetworkManager wrapper), ivy-pass (pass password manager wrapper), screenshot tool, ffmpeg wrappers for screencasts...

Many services have Emacs UIs like Slack (slack and ox-slack emacs packege), ERC (IRC&other chat client discord included), Jira (-mode), Maxima (imaxima), ESS for R development, various on-line wrappers including hackernews, various Google service wrappers, ix pastebin services etc

Try to look for "emacs $something" and you normally get some results, if you develop try to look for "emacs $language" and you'll find many options (also look on YT&c for thing like web-mode, emmet-mode, ...).

In the end I think Emacs is the modern and last living LispM implementation :-)

Personally, I like Word, because it has a particular niche - fast WYSIWYG use and mid-sized documents. That is, I can quickly make the exact nice-looking document I need, and print it. Without clunky UX (of Open/LibreOffice), and without it choking on mid-sized documents that's characteristic of GSuite and friends.

I would not call it WYSIWYG. Send it to a friend with a slightly different version of Office or Windows, or without the same fonts installed, or with a different locale, or a different printer, and it stops being WYSIWYG.

I'll grant you the version difference, but otherwise I strongly disagree. Font issue doesn't exist if you use standard fonts and is solvable in general by ticking the "embed fonts" checkbox. You're better off than with SaaS versions, which offer you a very small selection of fonts. With printers, it's actually MS Office that can paper over the differences between models and drivers; you're better off by default.

With the set of available feature, Word is as close to WYSIWYG as you can possibly get.

The entire 2013 Microsoft Office suite clocked in at ~45m lines of code, so Word is not even close to some of the really large programs considering Office also includes Excel, Powerpoint, Outlook, etc. [1] While we're on the subject, the biggest issue I personally have with Word is formatting, specifically dealing with images, tables, and padding. I feel that it's pretty satisfactory for day-to-day tasks, though.

[1] https://informationisbeautiful.net/visualizations/million-li...

Checking for passive voice is not something that should be done.

I can only assume you're asking because you're not using it for you day-to-day job?

Excel needs to die or find a better means of placing process on top of hideously embedded business logic that lives inside those sheets. Trying to reign in the entropy of VBA macros across an organisation is a _massive_ headache.

> We can build better bridges than we could a century ago, but there isn’t an incentive to replace a bridge until it falls down

WRT. Emacs/vim vs. VSCode/Atom, I can only say that we may know how to build better bridges than in the past, but for some reason we opt not to. It's like the goal of new software is everything except empowering users. Optimizing for ease of onboarding, for looks, for extracting value from users via business model trickery, etc. But not for utility.

Some other important factor about the software you have listed is that they are actively maintained, either because they have a community or a profitable enterprise behind them, I don't think inertia alone is enough.

Well... Few time ago I discover my "definitive" productivity app: Emacs. In few months it became the center of my IT life. And I still have to see ANY more effective, simple, flexible, pleasure to use modern application that can only compete with at least some aspects of Emacs.

The problem IMO is that in the glory time of IT software was not a business, was a free thing that came with hw and if it was good and flexible people are more willing to buy your iron. Now (not from now) that software became a product commercial need destroy freedom and evolution creating a modern, colorful, limited and limiting digital middle-age that can barely stand like Babel's tower...

IMO sooner or later tower will collapse and the more we wait the more it will hurt.

More about software: our society mostly live on text: laws are text, newspapers are text, books are mostly text, text is the cheapest and effective means of communication we have so if we really want to being productive we need to have something focused on text. Yes, we have also images, videos, audios, but they are still marginal. We have excellent audiobooks, super-nice videos and diagrams/images that communicate something an order of magnitude better than text but they are still hard to make, hard to design, rare, not really flexible etc. so while we have to work on them hard and constantly to better support/integrate our focus must still be on text. For many years to come.

Try only to do most common task without a keyboard and see how unproductive, limited and hard they are even with the best touch UI we have, with all "AI" support possible.

That's a lesson nearly nobody want to learn but that's also the reality we have to face, willing or not.

While I like to see competition, I still think Microsoft has the market for the foreseeable future because you get so much value. My yearly subscription for Office 365 family plan is just renewing and I can;t imagine not paying $99/year for: everyone in the family gets 1 TB of cloud storage, web based apps (which work just fine on Linux), and always up to date versions of the Office apps, if you want those too, and quality mobile apps for access to documents anytime.

I use markdown and Latex to write my books, but I still find a lot of value in Office for occasional uses.

I’m disappointed to not see our Dropbox Paper in that writeup, although it’s a good article! Wish they wouldn’t call it “word processor wars” though since that of course misses the point - it has too many associations to the printed page whereas I think everyone incl the author and the people quoted knows it’s about a more collaborative content creation experience.

Yes. Dropbox Paper is fantastic.

One advantage of Microsoft Office is that I can reasonably expect that in 20 years Word will still be around and will be able to read the files I write today. With these apps, I would be surprised if in 5 years I could read the files created with these apps.

Agree, data lock-in is probably the biggest concern I have with any new productivity or note-taking tool.

It's tough because innovations in tools generally require some degree of schema or file format customization. I don't know about other tools but at least with Airtable I know I can always hit the API to get my data, though I will miss out on certain business logic around it like data validation, and probably a lot of other details.

The best proxy I have to allay some of my data portability concerns is how much funding the companies have. Funding is a double-edged sword but comparing two companies, one funded at $50m and the other at $5m, it's hard not to go with the former. I am heartened though to see that some of my favorite tools (Airtable and Coda) have gotten very large investments, which tell me they'll be around for a while. When I see similar products not getting those investments, it's a strong signal to stay away or only use them as a toy.

New tools need to signal their trustworthiness by exporting to legacy formats. Even if it means losing something in the process. Give me CSV, DOCX, or an SQLite DB. Not exactly rocket science.

Or better yet, give me a native app that I can download to my machine and use forever no matter what happens to your startup.

The reason word is gonna continue to exist long after all these fly-by-night productivity startups have gone under or been bought up and wound down is because it doesn't rely on the cloud to work, though ms seems to be trying to head in that direction unfortunately.

> It's tough because innovations in tools generally require some degree of schema or file format customization.

It's not tough. All it needs is to allow me to download the data in whatever format you actually use to store it, or some serialization of it. Myself, or other interested parties, can reverse-engineer it if they're too lazy (or too much into abusive business models) to provide documentation of the format. It's simple as that. The lock-in isn't in the format, it's in the inability to get the full document (and possibly reimport it later) in any format at all.

It's worth noting this is a testament to Microsoft's (extreme) dedication to backwards compatibility.

I sometimes wonder who's more fanatical: the glibc folks or MS. But it seems to have paid off for both teams in the end, so (especially as a young engineer) who am I to judge?

This is especially noteworthy because the older Word formats were essentially memory dumps of the data structures used to store the document.

The binary formats that immediately preceded the current OOXML were OLE compound files (aka "structured storage), which is basically a filesystem-in-a-file that's intended to be used for serialization purposes - to allow files to have arbitrary nesting of COM components.

Individual data structures are then binary-serialized into that compound file, but I don't think it's accurate to call it a "memory dump". People often get that impression after looking at the compound file and seeing garbage there, much like unused blocks in memory - but that's because it's a filesystem, and as such, has a concept of unused "sectors". This is also why a freshly saved binary Word file might still contain bits and pieces of old data and metadata.

That's how the DOC format in the 90s and early 00s worked. I was under the impression that the early formats for Word on DOS, Windows and Macintosh were all different and represented in-memory structures written directly to disk. Unfortunately, I can't find a citation, so I may be mistaken.

There was certainly some format that Word and Excel used before compound files, since COM Structured Storage only appeared in 90s, and the first version of Office to use it was (IIRC) Office 97. That older format may well be some kind of a memory dump. But I don't think there are many Office files still floating around these days, and most third party software that works with them seems to assume that it's 97 or later.

What files?

When talking about SaaS (or more accurately, SaaSS[0]), you can "share" stuff between devices and people, but you don't control the data. You don't get a self-contained file that you can store and use independent of the service.

Just that is enough for me to prefer Word or OpenOffice whenever I can (and really, I just use Org Mode whenever I can get away with it).


[0] - Service as a Software Substitute. See also: https://www.gnu.org/philosophy/who-does-that-server-really-s....

While we have no plans to go anywhere at Notejoy, we do acknowledge that this fear is very real. So to address it we support full bulk export of all your notes to Google Drive at any point, so you can rest assured that you'll always be able to take them with you.

Don't take this the wrong way, but... you've flipped multiple startups. Why would I believe that "you have no plans to go anywhere" this time?

> So to address it we support full bulk export of all your notes to Google Drive at any point

Could you instead provide an option to bulk export all my notes to my own drive, via a .zip file?

The irony is that Google will probably shutter Drive before you guys go anywhere :)

Try my HTML-NOTEPAD then (https://html-notepad.com).

Documents there are just HTML files that will be readable for centuries to come.

What a great solution! I've been on the lookout for a basic (non-bloated) local, html-based word-processor to replace RTF (for myself and as a recommendation for others). HTML-notepad is loading very quickly on an older dual-core 1.2ghz laptop I'm using right now (xfce, debian 64bit). The resulting html is remarkably tidy.

Love the fact that it's portable! Plus... no data-mining:)

I have a large set of local html pages that I use for bookmarks, PIM stuff, etc. It's my goto personal organizer. Mostly html 3.x, lots of tables. HTML-notepad has handled these pages without issues. To my surprise, it even displayed a search form embedded on one of these pages elegantly. Really like the way you implemented the editing of links to minimize the steps: click on link, pop-up to edit.

Also integrates into Firefox as a "view source" editor - nice (since it handles my local html pages so well)!

The only basic tag that's missing is the HR tag (horizontal rule). It displays fine but doesn't appear on any of the menus. Tables are very well behaved but... minor point: I expected that pressing TAB within a TD would jump me to the next TD.

It doesn't seem to care about html tags it doesn't understand - basically leaving them alone. This is pretty awesome, since I use the following a lot:

      Title (data)
      Expanded, multi-line (data)
I've been "playing" around with html-notepad for about an hour and so far, no issues whatsoever. It's not often that a new program appears that could potentially become an integral part of my daily workflow. Thank you for sharing this!

Yep. A few years back I tried out a handful of new apps, or at least new at the time. Two have folded, the third merged with another service and offers very basic backward compatibility. While I'm not a huge fan of Office, and I think Google Docs/Drive is in need of a big update, they're both likely to be here in five years.

Not mentioned in the article but WorkFlowy (YC ‘10) is still going strong. Minimalist web-based word processor with a flexible outline format. Can hold hundreds of thousands of items, without you getting overwhelmed.

The same author actually wrote about WorkFlowy a few months back: https://www.geekwire.com/2018/workflowly-founder-new-list-up...

I'm a Workflowy user but would never describe it as a general-purpose word processor, even with the "minimal" modifier. There are no formatting tools whatsoever. It's more of a minimalist mind-mapping tool.

Good insights on the trends in word processor markets. But, including rich text editors and note-taking apps into the word processor market is bit of an overstretch.

Note-making apps are typically for digital screens, with a limited set of tools for document creation.

Word processors cater both for digital screens and print media. They'd have a richer tool-set and are for power document making people.

Interestingly both categories have made lots of improvements recently, when it comes to sharing, reviewing and co-editing.

Take, Zoho Writer (https://www.zoho.com/writer/) for example. The product beautifully carries the ease-of-use of simpler apps, without compromising on the richer word processor features. This takes the game one-level up. Simple writing apps now have to bring in more document creation features to compete in the market with word processors. Otherwise, they are just their own category with their own market.

I'm curious what open source options will crop up in this space. So far I haven't seen anything promising that compares to the Notion or the Air Table approach.

Notion and Coda are attractive, but I'm wondering if the learning curve is too steep. Not sure if my team would use them without me setting everything up.

I use Notion for personal projects and love it. Very easy to learn. I used it to plan and execute my NaNoWriMo challenge last month and really have nothing bad to say about it. Works well on all my devices.

But it's unstructured. I think in a work environment it's most useful as a wiki/project planner/lightweight tracker, but still can't replace the MS/Google office suite, and can't replace Jira/Asana for in depth ticketing.

I’d agree: Notion is great for individuals, a household, maybe even a small business? But it’d be a tough ask to scale to 100 people in an engineering org. But who knows, maybe it will find a niche like Basecamp did?

It is a wonderful replacement for Evernote though, and it makes you wonder what Evernote has been doing all these years? The formatting, drag/drop, multiple views on the same docs are great.

Evernote started the trend but Microsoft destroyed them in the same way they killed Netscape; OneNote is free beer and it's deeply integrated with the Office suite.

I can scan a whiteboard using MS Lens on my phone and it's instantly available on my laptop in OneNote. Oh and I don't have to convince the SOE nazis to allow install.

I can do the same thing with Apple Notes. And I don’t have to pay Microsoft anything. And it’s all available to iWork. Notes even has document scanning.

With Continuity Camera in Mojave, I can take a photo from my iPhone and instantly place it within a Pages document.

For power Excel users, Numbers isn’t yet a replacement, but for everything else, the stuff that is built in to MacOS works great and is free while working seamlessly across all Apple devices. I don’t work with massive spreadsheets nor use pivot tables so for me, iWork + Notes solves office productivity pretty darned well, plus Keynote puts PowerPoint to shame.

> I can do the same thing with Apple Notes. And I don’t have to pay Microsoft anything.

I don't understand what this means. I use OneNote and have never paid Microsoft anything.

Regardless of what other tools I try, I always end up coming back to Apple Notes for the ease of writing, drag & drop images, easy sharing as PDF, and being able to work on them on the go with iPhone and iPad. Apple's integration got me addicted. The search feature that can search even my sketches - bonus sweetness!

I do wish Apple Notes had some way of making a Kanban board.

Interesting, haven't heard of this one. From a quick perusal of the site it looks kind of like a more feature-ful Workflowy, is that a fair impression?

I'm not very familiar with Workflowy. They are similar though. I would say that Workflowy is what you'd get if you restricted Notion to only allow list items at the top level.

In Notion every paragraph is a "node", and a node can be text, media, list item, database (excellent feature, though, really more of a table where each item is a page), reference to other object, etc. Anything can have a parent of anything else, and the UI/UX is smooth.

For my NaNoWriMo project I've got databases of characters, places, maps, etc, and then pages for chapters with all the text, comments, to do lists, etc.

For the benefit of anyone else wondering the same: I checked it out and no, it's really nothing like Workflowy. The key feature of Workflowy is that you can 'zoom in' to any list item, essentially meaning any data item can also be a document which makes it actually viable to do everything at the top level. It doesn't seem Notion supports this workflow. Definitely jealous of some of the advanced data-type features in Notion but it feels much too fiddly for my mental model. Thanks for the explanation, though.

We've tried to focus Notejoy on being as intuitive as possible to support any kind of team. Because of this, we've seen a variety of teams beyond just startups adopt Notejoy, including coffee shops, law firms, freelancers, agencies, coaches, and more.

My problem with these sorts of tools is always the same: I have been spoiled by my emacs keybindings ( evil mode ), the ability to parse structured text, org-mode, and command line tooling.

I do not have to open a web browser and navigate, I just launch my helm-project-search buffer and type `org/work`. I then have a fuzzy search that matches on file name. Toss these files into a dropbox folder, and you are good to go.

None of these applications I have seen support structured text such that I can export / import through Emacs. They cling to your data as tightly as possible and do NOT want to let it go.

I was working as a hobby on an RPG supplement for a popular anime series. I was attracted to notion and coda, but was dismayed to find I did not have a way to turn the data I entered into structured data which I can query to generate an attractive webpage for the game.

I'd love to have one notebook-type app that works between all my devices, or some combination that lets me have the same or similar content.

I loved OneNote back in Uni, though I reached a point where I stopped taking notes (through the fault of the powerpoint teaching method, and my own laziness), and OneNote does seem to be a bit bloat-y when it comes to sync.

Stuff like EMACS org-mode (or, preferrably, some kind of Visual Studio Code equivalent) seems enticing, but I'm not sure how I could hook that into my phone in a way that isn't a terminal emulator.

Anyone found anything (app, service, or custom setup) that works for them?

I use orgzly to take notes in my phone. It can create and edit org files and sync the files through Dropbox. Then you can edit the files in your computer through a Dropbox sync folder. It's just a nice app for taking notes for me though. I haven't even edited the notes through my computer. Good to know that my notes are backed up without having to use some opaque backup method.

Try out Notion. Works great on all my devices (MacBook, Chromebook, Android). Free tier is quite generous.

I really like that in Notion you can just spin up quick data tables with rich data objects underneath.

But also -- don't underestimate just having a notebook. For daily todolist type stuff I just use a pocket sized moleskin and the "Bullet journaling" method.

Yes, there's a bunch of org mode client that works everywhere, I built this one: https://github.com/mickael-kerjean/filestash/wiki/Release-0....

OneNote's sync has been completely rewritten, if you haven't tried it out in awhile, give it a go. It is much faster now, literal orders of magnitude.

There was a Firefox thread on here yesterday. Interesting one.

My take was that 10-15 years ago, FF's feature gap was enourmous. Tabs. Extensions like firebug for web development, adblocking for browsing, various "social add-ons" that were innovative and exciting. IE 6 was a dinosaur.

The question for these is.. what more do you/we want from a word processor?

Integration with documemt and version control with anything other than sharepoint. Something that is guaranteed to work without application crashes. Configurable Cloud storage/saving. 100% open document formatting to allow for interoperability.

I thought Wang made the first word-processors.

Applications are open for YC Summer 2019

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