Microsoft Word was never designed to be an ideal tool for creative writing, programming, or any such thing. It was designed to help office workers produce business documents (a) quickly, (b) in a format that integrates with all the other programs in the Office suite, and (b) in conformance with whatever formatting and workflow requirements their employer already had in place. Microsoft probably consulted with a bunch of Big Business customers when they designed Office. Yep, the kind of Big Business that uses Java classes like EnterpriseBusinessInterconnectInterfaceFactoryFactoryFactory.

Type a few word, hit "Save", and automatically get reasonable default fonts and margins? Check. Type a few more words, make some typos, have them fixed automatically? Check. Certain words need to be italicized or underlined? Check. Certain words need to be in a different font? Check. Who cares if it's consistent, the boss wants it bold so just make it bold. Indent the first line by X inches, double-space here, single-space there? Check. All accomplished with a few clicks of the mouse. No need to learn any command-line programs, formatting \Syn\{TaX}, or keyboard shortcuts. It's exactly what the majority of office workers need. Bonus points if it also helps clueless parents design their daughter's birthday party flyer in pink and purple Comic Sans, but I don't think MS really cares because that market is miniscule compared to Big Business.

> I hate Microsoft Word the way Winston Smith hated Big Brother.

At the end of that novel, <SPOILER> Winston loved Big Brother. </SPOILER> Because he probably realized that no other program but Big Biz MS Word would fit the use cases that it was designed for. LibreOffice? Call me again when it gets the Review function right. Your average college professor isn't going to learn how to use a version control system to suggest changes to his student's thesis-in-progress.

Word has version control though, and it's appreciated by many in academia. Version control for humans is a Big Startup Idea, I wouldn't be surprised if Dropbox and Github were trying to tackle it (from 2 different angles: Dropbox from the "how do we get average the user to get features for nerds", and Github from the "how do we get the average non-nerd to use us")

Word's review functions appear as pretty colored bubbles and a bunch of icons on the Ribbon. It's easy enough even for Humanities professors to use on a daily basis, despite the fact that as a VCS, it's strictly inferior to something like git. Actually, I think it's popular precisely because it ain't git. Most professors who are fluent in Word's review functions would be hopelessly lost when faced with a Dropbox conflict, let alone merging git repos.

As for git's complexity, it's all in its completely opaque display of its internal state. A chain of deltas - diffs - isn't complicated. It's all git's (perverse) desire to follow the exact-same command-line interface as cat and awk, when neither contain persistent state.

Word shows pretty bubbles next to sections that have been edited. The bubbles are color-coded so it's easy to tell who did what. The edits can be accepted, rejected, or further modified with a couple of clicks. There is no list of commands or keyboard shortcuts to memorize. Just a bunch of icons, menus, and clicking around.

Word doesn't require your content to be made up of neat lines of verse, either. Natural languages don't organize themselves into 78-char lines. It should be OK to edit a few words in a long paragraph of prose. Even Wikipedia's messy "Revision History" is probably better than git in this regard.

It is a horrible representation for stuff where edits may consist of both typo fixes, and separate wording changes in a single sentence - where you might want to accept change to one word and reject change to another word, despite that they came from a single "commit" to a single line.

For collaborative text editing both word's track changes and google docs features are far superior to git.

We're bringing version control to people who collaborate by email, starting with Gmail. All people have to do is exchange drafts by email (like normal!), and we infer the history, as if they've been using git.

We're looking for beta testers:

https://draftable.com

https://draftable.com/gmail/faq

with the answer to your question (and more), but I'll answer here too:

We don't modify the data in your Gmail account, so if you try Draftable, then uninstall it, your Gmail account won't be altered. Furthermore, if you disable Draftable before uninstalling it, we'll delete our index and revoke our access to your account.

It's meant for people who exchange documents by email.

Collaboration in physics research: hand-written notes on margins of said pile

The issue with git etc is that everyone involved has to understand how it works and accept and appreciate the benefits over the system outlined above (which works quite well for papers with few authors). If even one author doesn't care, you cannot implement the system. And good luck getting professors to care about things that aren't research or funding.

http://math.andrej.com/2013/06/20/the-hott-book/

And if some of them can't use it, teach them. And revert to the previously non-fucked up version (it's on someone's email by the name of Article_1_fixed_2_corrections_10_10.docx)

Something I like about version control is that it at least ensures continuity of the version.

Yours sincerely, a graduate student whose thesis is written in LaTeX and then converted to RTF and then hand edited so his supervisor(s) can give him comments.

I sent them a link to my github repository, but they never took me up on my offer :(

I've found that collaboration on documents via Markdown is not too crazy bad. Otherwise, I ask for comments on the PDF or on hard copy. Going back to latex from RTF is an exercise in madness.

And I second the pandoc recommendation if you can get it to work

The worst part was that they didn't even use the track changes feature on anything I sent them.

I actually do make bit of things in LaTeX through Org-mode at the office, and sending things in PDF actually makes it clear what kind of handling I'm looking for. If I'm actually expecting someone to revise my writing, then I'll just send them plaintext.

Now considering even Adobe Reader allows commenting, I don't think there are many of issues any more...

I don't know the state of visual tools for git but tortoisehg for mercurial is quite easy to get started with.

After forgetting to push after commit and then dropping your laptop, you'll know.

Nope, like regularfry rightly pointed out.

>The only difference is that you still can use your local git repository if your server burns down, but you can't do that with svn. Even then I can `svn relocate`.

Look, I am not defending SVN for software development, I am a happy git camper. But git for technical writing is simply the wrong tool for the wrong job.

I have had various beginner git users get very confused when dealing with conflicts -- I have got myself quite confused on occasion too.

1. http://www.wired.com/wiredenterprise/2013/06/cades-witty-hea...

Mostly because he was drugged, physically and psychologically tortured and reduced to his basic instinct and given option to experience its greatest fear or 'love' Big Brother. That's closer to real world than it being the only good option. LibreOffice needs a good compare function, not the whole review shtick.

Sounds remarkably like using Microsoft Office.

Basically, you pulled this out of your butt because it hit the right notes and because it fit the preconceived notions of the readership of HN. WYSIWYG word processing was actually pioneered at Xerox PARC and MS Word has a direct lineage from there. Fitting the office correspondence conventions of the time was the obvious place to start. In a world that still operates off of paper, this is what makes sense at first.

Java classes like EnterpriseBusinessInterconnectInterfaceFactoryFactoryFactory are a direct result of the deliberate crippling of the Smalltalk runtime model that Java was based on. The very small and minimalist Smalltalk runtime model was another product of Xerox PARC, BTW.

Back in the day, even Java was one of these subversive back-door hacker/engineer things management wasn't hip enough to be a part of. (Unless you were one of the even hipper folks into Smalltalk, Dylan, Common Lisp, Eiffel, Tcl/Tk, Perl, Python, what have you, in which case, you knew better than to fall for Java.)

Heck even SQL, Fortran, and COBOL were the new hip thing back in the day. (Even if that came with corporate backing.)

Actually, they are a consequence how easy Java allows to write such type of code and how enterprise architects design software.

I have seen EnterpriseBusinessInterconnectInterfaceFactoryFactoryFactory examples in the enterprise done in lots of languages.

The same way there are developers that write FORTRAN in any language, there are those that make Skyscrapper designs in any language.

Then indeed, its influence over those fields should be allowed to wane.

LibreOffice? Call me again when it gets the Review function right.

I was required to use the Review feature as a sort of half-assed collaborative editing system for tech specs on a couple of projects around 2008-2010. I found it to be a buggy, work-destroying trap of a feature. Things would actually have gone more smoothly for my project Microsoft had never implemented this feature.

I haven't tried any similar features in LibreOffice. If it doesn't do anything totally crazy (like randomly duplicating blocks of text in odd places) it's probably better.

I read an interview with some high-muckety-muck in the MS Office group years ago. He actually said, in response to a question about features vs. bug fixes, that he was sure users preferred new features to bug fixes for old ones. That features got people excited and no one really cared if they lost a little work every few hours.

He didn't say he'd asked any users. He was just sure.

I wish I could link to it, but I don't recall where I read it.

He's probably right (if you think of users as being the person who approves the payment). I can't imagine the CFO at a Fortune 500 company approving a few million dollars to be spent on bug fixes. The obvious question would be "why did we pay for defective software in the first place?".

So perhaps one side effect of the subscription model will be less buggy software, since the business model relies more on keeping users than getting them to upgrade.

Don't bastardize the message and theme of the book.

  IGNORANCE IS STRENGTH.

That's Abiword. A much smaller and cheaper piece of software.

That is all that most office workers need. So why is Office, a huge piece of software the default?

Abiword's grammar checker is not nearly as good as Word's. That is specially in non-english laguanges. Also I would add to that list all the templates and design that Word brings out-of-box.

I am a weird case because I use Abiword at work and Office at home. And at least for short documents (10 pages max.) the experience on Word is just more enjoyable to me.

From a linguist's point of view, the language that English is grammar checked against in Word has always been a broken approximation of an uninformed prescriptivist's version of English.

An example opinion from the guy who wrote the book on English grammar:

http://itre.cis.upenn.edu/~myl/languagelog/archives/005061.h...

I have actually only used one working spell checker for Swedish: stava. And the ideas behind it are trivial, so I guess that nobody else has written a specialized spell checker for Swedish. Generalized ones like ispell, aspell, hunspell and Word's do not work for Swedish.

http://www.nada.kth.se/~viggo/stava/

LOL seriously? Look everybody, v2 even has footnotes and -gasp- tables! And they are 'very powerful' tables, too, because you can even merge cells (in a 'non-modal dialog', no less!)

(oh, and last release version - 2009).

Most users don't care so much about the "size" so much as the cost.

You both got it wrong. It's not about size. Not about cost. It's about compatibility. People use word because their boss, publisher, or professor uses word. Because an entire ecosystem of office software integrates in word. Because kids taking a computer class in school learn Word commands, and nothing else. Because a perfectly styled document in LibreOffice doesn't always look right in Word.

More than 20 years later, Microsoft still survives off the popularity of DOS on IBM compatible pc's in the late 80s and early 90s. Their strategy since has been to further entrench themselves into our daily lives.

I suspect this is the ultimate longevity of Word. Retraining is a real pain. So the only way to kill Word is by a thousand cuts. One of those cuts seems to be Google's documents. A number of non-technical folks can now send me a document in Google Docs so I don't have to switch from my Linux desktop to something else to access it, and Google has done a reasonable job of auto importing Word.

So domain specific solutions, with "good enough" import/export may be the solution.

Unfortunately as Publishing is now having to deal with online and electronic it means they are not best placed to deal with the changes.

And book publishers have sclerotic processes (makes teh Laudry service seem slick and efficient) to take a book from delivery to publication and extra month or so playing with word is't going to make much difference

And that's why they released a "Home and Student Edition".

Student Edition: "Give them Office now at a discount, and they'll keep using (more expensive versions of) Office when they get a real job."

Office is named Office for a reason. Its entire existence revolves around the typical 20th-century white-collar office. All other use cases are secondary.

Just a thought.

They prohibit you from making money with it? That blows my mind. Not only is it completely unenforceable (the PR backlash from Microsoft doing an enforced license audit of the likely purchaser of that version would be immense), it's completely pointless. Again, the products are differentiated by what they lack.

I wonder if the Microsoft legal department counts creating lemonade stand or yard sale signage to be "commercial use"...

However, I don't think there's a way to upgrade if you buy Home & Student and then discover that you need to do commercial work with it. You have to buy a new license.

By the way, Windows includes a free word processor, WordPad, that does what most home users need without them having to buy Office. There are also free viewers for Office docs...

Microsoft Office is commercial software costs roughly $300. If you want to use it for commercial purposes, you pay the price. However, there's a special deal where you get a Home & Student version for a third of the price, on condition that you don't use it for commercial purposes (ie as a replacement for the version you should be buying).

A lot of commercial programs are available in cheap student versions or in free versions for home (non-commercial) use. This is standard industry practice. It's not a Microsoft thing.

Confluence? Twiki? Or equal such centralised documentaiotn and revision service, be it SaaS or locally hosted, Yes.

He pointed that out. And he pointed out that the industry was still foisting it on him. That is why he wants it to die. This was the summary at the end of the article in fact.

I don't think most people distinguish "Word" docs from any printed doc, except those sophisticated enough to use "Excel" to produce columnar-format data ("Word" tables are hard to do and sucky looking).

I've been using it for years now and it does almost everything I need it to do. I realize there are people who absolutely need every little feature in Word, but for things like letters, technical reports, briefs or software documentation Pages works just fine, and produces nicely-formatted documents. You need to ignore the Apple marketing, for some reason they think Pages and Numbers are used exclusively in a home setting for producing toy documents.

I just wish Apple devoted more time to Pages and Numbers, because the tools become annoyingly slow with larger documents (larger meaning a 60-page report with tables). I'd much rather see the existing tools optimized and working fast than new features.

And yes, I know this is not a perfect solution. I just think it's better than Word. But I will also point out that LaTeX (or plain TeX) isn't a good solution either. For people who don't know it well, it doesn't produce the results they want. And for those experienced with it, it becomes an unbelievable time-waster because you spend inordinate amounts of time tweaking things for no good reason.

As an aside: Did you know that you can create a halfway decent xhtml from Pages documents by exporting to epub and extracting the embedded documents? I've used that in the past to create a simple database my wife can edit using the tools she knows (Pages) by converting that stuff to an xml db. You can even script all the intermediate steps using AppleScript. As a platform to hack on, OSX can combine the strengths of Unix with the UI scriptability of Windows OLE.

Microsoft Word is also a styles-based tool, and produces nicely-formatted documents that are consistent with a stylesheet.

The problem is that people don't use the styles.

I only needed to style H1, H2, P, EM and possibly A (they don't have those names in Word, but you get the point). That's probably the simplest document structures you can have for a task that reasonably makes use of styles.

The biggest problem was, those styles are like liquid, slipping between your fingers as you work with them. You do one thing to the text and suddenly you find yourself in a style that's "adapted" from a style you've already defined, I think I've even encountered a new hybrid style between two of my user-defined ones. And then there's of course the "built in" paragraph and heading styles, occasionally thrown into the mix as well.

The point is, this is not at all like an actual style sheet based way of working, such as with CSS/HTML. You could get the above result by using (bad) code such as <h2><p><em>OOPS</em></p></h2>, you'd also get a mix of three styles. But the thing is, in Word, all these new hybrid styles got added to the style sheet!

So get this, I wrote part of a document, decided (as long as I have to do this in Word ..) I'd use style sheets, made the styles, continued writing on the document and the existing styles changed as I moved bits of text around.

Not having exactly that happen, is one of the main reasons to have a strict separation of content and style in the first place, is it not?

I'm sorry but, I don't think Word actually supports "styles". It's got some stuff that kind of looks like styles, but once you actually try to use it, it'll just waste your time at exactly those moments where you'd expect a proper styles system to save you time.

At some point I had finished the content and I decided to make the H2 a bit bigger font. With styles that should be the easiest thing right? Haha! Right. That's when you get to see all the adapted styles that suddenly appeared in your H2's markup, and they don't change along. So instead of just changing your styles, you are going to have to edit your document content anyway.

The whole experience was very, very messy. In hindsight I'd have preferred to keep the "style sheet" in my head and apply all the markup manually, like most people do, instead of relying on the program to take care of this for me, but wasting my time instead. No wonder that's the route most people seem to take.

(disclaimer: this experience was a few years ago, maybe they fixed it by now and styles are actually styles and don't change while editing your content, and are in fact either applied to some part of the content or not, and not somewhere in between)

I'm a LaTeX addict and not an HTML guru, so I'm sure there's a reason this solution doesnt work, but it seems intriguing.

Although, a few years ago, a couple of friends of mine were working on their masters theses in non-computer related fields (literature and biology), they complained about Word and its tendency to get somewhat "wonky" when working with very large documents (I am glad I have never had to experience such a task). I talked to them about the basic ideas behind the LaTeX workflow, that writing the text and doing layout/styling are straight from the start intended to be two separate concepts--you're either working on one or the other. And how the text content is also just that: a plain text file with a few document structure-related codes in it. A plain text file doesn't lose whole sections or mangles up chapters for no apparent reason.

I was surprised to hear their reaction, they were very intrigued and would have been readily prepared to learn the basics of LaTeX, if it meant getting that peace-of-mind: that writing a large chunk of text means that it is there, without a program doing behind-the-scenes modifications on it if you decide to move parts around, or so. I say "would have been" because they were already quite far in their theses so it would've made little sense to make the switch still.

Because that's really the point, isn't it? That's what annoys me about Word (and a lot of other word processing software btw), it just won't keep its fingers off my document, and all sorts of things happen under-the-hood. With LaTeX (and similar solutions, perhaps Markdown) the whole workflow is different: first you write the content, with some minimal markup and structuring codes, and that's a text file. It's yours and nobody, no software is going to touch it. It's source code. And only then you apply a program to it, in order to compile it to a display format. But that program will not touch your source.

I really like it that way. I understand it's not for everyone on the consumer market, but it's interesting to consider how different that paradigm is.

You'd have to create quite complicated style templates to get HTML formatted to fit onto pages.

Pages has always scaled fine for me, however Numbers will choke on a gigabyte-sized CSV. I tried to, import the US census data into it. I couldn't get Numbers to work. It crashed every time. Microsoft Excel could just about cope but was struggling to scroll or filter data. Eventually I gave up and threw it into MySQL. The right tool for the job.

Although Excel has since been corrected for this use case[1], it isn't even the right Office tool for the job. That's what they created Access for.

[1]The SQL Server team built PowerPivot for Excel 2010+, which basically imports your data into a SQL db that gets embedded into the Excel file and transparently queried from Excel proper. Runs smooth as butter, and pacifies the Excel purists who refuse to lower themselves into using Access (I've worked with several of said purists).

Writing code in Word is completely possible, but would you fault Microsoft for the ensuing bad experience? You can also use Eclipse for writing a resumé, but would you fault it for not supporting that use case very well?

It means you are forced to consider buying an expense Enterprise SQL license once you are using it in earnest. If it wasn't for Excel chocking on 5million rows I'm convinced most businesses wouldn't bother with databases at all.

Your comment is a bit like saying "Most businesses wouldn't bother having a cafeteria if only Excel could cook meals."

Excel was never designed to replace a database and the fact that it can't isn't some evil plot by MS.

This part is tongue in cheek, but true. There are businesses out there, where excel is used for everything, including many things it should never be used for. Seriously powerful app. Seriously (ab)used.

I don't hate Excel, but I hate the "craplications" hare-brained power users create.

I've found the same to be true for editors. Most text editors have struggled to open large log files (500+ MB) whereas cat can open a file of basically any size.

Of cause, for parsing logs a few well thought out awk commands combined with sed, sort, unique and other *nix utilities usually beats everything else in my experience.

The day will come when you will need those two figures side-by-side, together with a caption positioned just right. Or a table that spans multiple pages and has multi-column cells. Or, heaven forbid, you will need to submit your bibliography in a byzantine format invented by someone with nothing better to do ("we take pride in the fact that we place a period after author's names").

LaTeX also has another problem: even if you know it, getting from a document.tex file with 0 bytes to a document that looks good takes much longer than it should. By "looks good" I mean doesn't use Computer Modern, and overall doesn't look like a thesis from the 70s.

Not sure if you mean to say it is difficult to do or not in LaTeX.. but I had to do just that last week (10+ pages long tables, side by side figures with fancy captions, etc.).. and it was just a matter of searching for "LaTeX $problem" and I was in business..

To be fair, I would advise almost no one to do the same. Something better is needed[1], but I think that most potential competitors know better than to attempt to challenge MS Word.

[1] And, indeed Google Docs is usually passable and offers some advantages.

I've never understood this way of thinking. Either it works or it doesn't work -- the age should have nothing to do with it.

  \item   
  \begin{circuitikz}[american voltages]
  \draw 
    (0,0) to[battery, l=12V](0,3)
    to[R, l=$150\Omega$ ](3,3) 
    to[D,l=$D_1$](3,0)
    to[short](0,0)
    ;
  \end{circuitikz} 

If some new technique/workflow/whatever is newer and more efficient, then I'm all for it. I just don't like the attitude that "newer == better" that has seemed to become more and more prevalent in recent years. (I'm not saying you were doing that, but the phrasing you used is something I typically see in that situation.)

I think LaTeX starts marginally easy, gets complex and fiddly but then pretty much just stays at the same level of complex and fiddly. Word starts very easy, gets complicated quickly and eventually presents un-fixable problems, no matter what your skill level.

That is in addition to the horrors you unleash once something stops working and you need to dig deeper. dvi2pdf..

These tools play well with git and make it easy to automate an analysis. I am a strong believer in literate programming - embedding the code in the documentation where feasible. There are case where it is not (long Monte Carlo simulations) but I can at least pull in the results and have the source code that generated them in version control and the report automatically uses the latest version.

Microsoft kept changing VBA and each release would break things. We had third party tools that had locked VBA modules and ours just wouldn't permit processing data in these workbooks.

I can extract all the data with R and am much happier. The Open Source community seems to better support their products than Microsoft. Go figure...

So I don't know why the original article suggests competition is dead -- unfortunately Pages is not cross-platform, but Google Docs is certainly a worthy, simpler competitor (though perhaps too simple).

It has its quirks. It's also damned powerful and optimized.

(I guess that's including its program-ability and the ability to interface it with external functionality via Win32 et al. I remember, for example, referencing IE 4, then 5, in order to "add" regex's to it via VBA scripting. Actually, as I recall that was actually Windows Scripting Host; installing the relevant version of IE at that time in turn caused WSH to also be installed.)

P.S. I was not number crunching during the Lotus heydays. I gather its products had some features that were unparalleled for their time and that Excel may never have fully matched.

P.P.S. As for quirks, I remember one bit of oddity emanating from a formatting function that an older programmer associated with a Fortran convention, when I happened to describe it to him.

http://siag.nu/

(At least it's not trivial to get it to compile on a modern Linux distro as far as I can tell...).

Not that it's really a replacement for excel for all users, but it was an interesting package.

These are the limitations of Excel alternatives. They look nice on the outside, but fail to understand that users don't use them just for the simple stuff. It is well known that Excel is what really powers many companies, and these are often absurdly complex.

(1) OO Calc insisted on drawing an x-axis label for every single point in a graph, on every redraw.

LO Basic is not VBA. This is important when people's stuff is written in VBA.

I've tried Open / Libre Office on both Windows and Linux targets. A benchmark: Create a couple columns of, say, 10000 x/y values, and see how long it takes to create a basic scatter plot.

Also, I've noticed that the user interface of Excel is a lot more polished. For instance it's more forgiving of exactly where I put the mouse pointer before clicking -- valuable for me because I get severe eyestrain headaches whenever I have to look too closely at the screen for too long.

I’m by no means an Excel power user, but Apple Numbers provides everything I need in a spreadsheet app. I have native Numbers apps on my laptop, desktop, tablet, and phone (but there’s also a web app). Whenever I save a file on one device, it’s automatically added to all the others. A file created on my laptop can be edited on my phone and vice versa. Also, Apple Numbers imports and exports Excel documents. As a result, I haven’t opened Excel in over a year and I doubt I’ll ever buy another upgrade.

Excel may not be replaceable for everyone, especially if you’re using it in a corporate setting, but for me, I have found I can easily do without. I find Apple Numbers to be way more user-friendly and convenient.

Dropbox+Excel seems like it would do the same thing, but even with more flexibility (including version control, document sharing) and supporting even more platforms.

I don't think you can ever consider spreadsheet software to be 'user friendly'. They're spreadsheets.

Numbers is not just a spreadsheet editor, it’s a full-fledged spreadsheet creator. Its features are sufficient for all but the most demanding users.

> Dropbox+Excel seems like it would do the same thing

For me, it doesn’t. Just using Dropbox and my stock OS (and providing I don’t use Numbers), I can view Excel files on my phone and tablet but I cann’t create or edit them. I still need to use a third party app and Microsoft doesn’t have an appealing Excel offering for mobile platforms.

> even with more flexibility (including version control, document sharing)

Numbers has version control and document sharing.

> supporting even more platforms.

Numbers is available as native apps for iOS media players, phones, and tablets. There’s also a native app for OS X. For Windows, Linux, and every other platform that can run a modern web browser, there’s a very capable web app, available for free.

Excel is available as native app for Windows and OS X. For Windows Phone it has a very bare-bones app, the Android and iOS versions are a disgrace.

> I don't think you can ever consider spreadsheet software to be 'user friendly'.

Maybe you should try Numbers, especially on a mobile device, and then compare it to Excel.

Then again, I carry my air on me everywhere I go, so I'll never, ever be editing spreadsheets on my phone. Seems like an odd usecase to me, though.

I have worked with Excel-based trading systems with tens of thousands of lines of VBA code, controlling systems that trade hundreds of millions of USD.

If Excel stopped working today, by tomorrow the stock markets would likely be in a blood bath, such is the ubiquitousness of Excel in that field.

Having worked with some VBA myself, this sounds like a very, very, scary statement.

Apparently a reference to Former Duke Cancer Researcher Anil Potti. [1][2]

> …put egg on some Harvard Econ prof's faces - to the delight of the UMASS grad student that found it...

Likely referring to the Reinhart/Rogoff spreadsheet error discovered by Thomas Herndon. [3]

[1] http://en.wikipedia.org/wiki/Anil_Potti

[2] http://www.huffingtonpost.com/2012/02/13/anil-potti-duke-can...

[3] http://chronicle.com/article/UMass-Graduate-Student-Talks/13...

That something is powerful enough to do the job does not mean that it is a good tool to do the job.

Of course the particular use case I mentioned was probably pushing Excel right to the edge of the envelope and the company was actively developing a .NET replacement for it because everyone recognized this. I have no idea how that turned out but I can say that the Excel system was very impressive and very functional.

Each localized version stored function names as localized strings!!! So format was not even compatible across the same version.

https://github.com/python-excel/xlrd/blob/master/xlrd/formul...

Having problems is not the operative issue. It's userbase. All software has problems; Excel is one of the projects that has a large enough userbase that you can schedule conferences about its problems.

P.S. "EDA" is of course more than one tool, but it's close to the right scope and what jumps to mind right now.

Exchange as well.

And as one of the comments below said, try Lotus Notes for a while :o)

Notes has the capability to make anyone love Outlook.

Nowadays I would gladly have it back.

As a user in a mixed MS Office + Google Apps + LibreOffice environment, Google's recent work with QuickOffice has been EXTREMELY helpful. My big gripe isn't Excel so much, either, but people who refuse to abandon PowerPoint when all they need to do is create a bullet list.

No, not at all like those. Big corps like these: http://money.cnn.com/magazines/fortune/fortune500/2013/full_... When someone says "big corp" they do not mean "Jason's Deli".

Although the most annoying thing about Excel is actually the users who insist on using it like a database...

Are you sure? Does it not strike you as odd that OpenOffice/LibreOffice still haven't managed to replace Word?

That it has not for everyone or as a whole doesn't strike me as odd in the slightest. There are plenty of reasons to use Word. Aside from the fact that it's solid itself, it also has a metric ton of inertia. Those using it already have the license or pirated copy, already know it, have legacy documents using it, and are generally otherwise are left with very little reason to switch to something merely because it's a "solid alternative."

I haven't yet met a single person who is happier with OOo/LO than with Word; if that's true for you you're the first such person I've heard of.

Much was due to the ribbon interface, which I liked, which LO doesn't have. My mind reverted to menus via Office 2000 which I cut my teeth on as a new graduate years back.

There are some things I still find odd / bewildering:

Right clicking on a block of text I have the options of both 'Text' and 'Character' which have different options. Neither of the contained options contain vertical alignment choices. Tables just feel a pain, zooming in and out as different cells / borders are clicked. Blah blah blah other small things. For my sins, I like VBA (familiar with, rather than romance).

Familiar is the key. It took a small learning curve, and I'm as productive in it as a modern version of MS Office. Much of the learning curve is having been so familiar with MS Office, sometimes things are just done differently, mainly at the UX Level. Integration for a personal and work based workflow is where Linux leaves Windows in the standing-gates was the game changer for me, and why I put myself through the learning curve.

I do feel LO's UI simply doesn't look beautiful, indeed, I feel it looks ugly. Everything is gray, some icons have shadow, some not. I have the feeling a few hundreds of hours on grouping icons, highlighting contrast in dialogue boxes, would make it a lot less 'tired' looking. Mozilla did it, I am sure LO will, and that may be a small but important tipping point.

The point is not that they are better but that they are a suitable replacement for all but a very small niche of power users. But people tend to compare the limit cases even if all they'll ever use are the day-to-day functions.

> The point is not that they are better but that they are a suitable replacement...

Well then you're just confirming what I already said...

Hi, I'm one such person. Reason: the ribbon. In the rest, yes, Word is better. But I like standard menubar interfaces and hate the programs that don't let me use them.

Could you explain what it is that scared you so much? I am the one to scream whenever I'm forced into using anything from the post 2003 MS office suites, which is thankfully extremely rare.

- LibreOffice can't even show the simplest documents correctly on the same computer: http://i.imgur.com/0glr43F.png vs. http://i.imgur.com/g3a1332.png

- Word takes < 200 ms to start up on my (blazing fast) laptop. LibreOffice takes > 2 seconds. It's outright irritating to say the least.

- Everything else has lags too. File->Save takes > 2 seconds to show me the dialog. In Word it's instantaneous.

- No smooth scroll?? My eyes!!! >___<

- No "Styles"??

- The UI just screams "LINUX", feels freaking hacky, and doesn't feel the way it should. For example:

1. Word highlights text word-by-word; LO does it letter-by-letter. Selections never work the way you intend them the first time.

2. Highlight a piece of text from left to right and then press Left. The cursor is now still on the left-hand side of the previously-highlighted text in Word, but still on the right-hand side in LibreOffice. What the hell???

3. A million other things about the UI doesn't feel normal, e.g. http://i.imgur.com/HM52E9A.png vs. http://i.imgur.com/zTK1zoi.png

I could go on and on, these are just what I found/remembered in a couple minutes of using it again.

For the smooth scrolling, you need to enable it under:

Tools->Options->Writer->View

Unfortunately, it is disabled by default.

I love the idea of Free software, use LO at home, and pushed for it at work. After a pilot LO rollout, we ended up going with MSO 2010 because of LO's poor interoperability with our old Word docs.

http://www40.zippyshare.com/v/24126556/file.html

Try opening it in MS Word and compare it with LibreOffice and you'll see why I ran away screaming:

http://i.imgur.com/4tSWXsn.png vs. http://i.imgur.com/3tBLuff.png

2. Ah, I think you're right about the missing font, I didn't realize that. I'll get back to you if I find another example, there are a million things I've noticed over time that don't render correctly other than fonts. (EDIT: See my other comment.)

3. Actually, I do have smooth scrolling "enabled", but it doesn't have any effect whatsoever (as of version 4.0.0.3).

That's a .doc file, an un-/poorly-documented format. As noted in the article, it's not simple at all. It is a good argument in favor of Libre and OpenOffice, though: other word processors will be able to open your documents if you decide to switch from them. Just that alone is a big win for anyone thinking long term.

All of your other criticisms other than the speed stuff boil down to "I'm used to MS Word". Which is fine.

I find Word's attempt to highlight word at a time to be extremely frustrating and distracting. Very often in my documents it simply will not properly recognize the boundaries of what I want to select, and small increments of the mouse produce jumps in the selection that were not intended. I think OO's behavior is a lot better in this respect.

Use the keyboard if you want to highlight character-by-character.

Word is simply optimizing for the most-common case: copying-and-pasting whole words, sentences, or paragraphs. It's a lot faster when you don't have to hit the exact character boundary of the word -- especially if you're using a touchpad instead of a mouse.

Except this is the common case, so it makes sense to have it on by default.

Add a nonlinear solver to calc? Sure.

Add a fill function into draw? _Maybe_ a two-month after-hours project.

It took me around 3 months, full-time, to write an import filter for LibreOffice that still only kinda-sorta works. And it depended very little on the rest of the codebase.

I think you underestimate how complicated a modern office suite is, and how much years-old technical debt there is floating around in LibreOffice.

So, I have this project in which I need to fill an area in a drawing in a proposal for a client. So my options are:

a) Spend $200 (about an hours worth of billable time, or 2 hours during dry spells) on MS Office.

b) Spend 2 x 4 x 10 = 80 hours (which is no where near enough, but let's go with your number) on this feature, then spend again that on finding and convincing the right people to include my patch, on a half-assed implementation which does just enough what I need, and tell my client 'yeah get back to me in 2 months when I have patched my word processor to deliver the functionality I need'.

Choices choices, which one should I choose?

Just that time alone on a single tiny feature would pay for a lifetime worth of MS Office software!

If you need/want to insert various objects in various documents (eg: graphs from spreadsheets in documents that dynamically update when the data source update) -- not so much. I could argue that you'd be better off using literate R/LyX -- but unfortunately there's a bit of an UX gap).

I much prefer LibreOffice to MS Office -- but mostly because I don't feel quite as a afraid that my documents are going to go randomly get corrupted at the worst possible moment without any reasonable hope of recovery (documented formats ftw!).

And if you really want to go down that rabbit hole of perfect compatibility, it's an absolute clusterf*ck. There are more than than 20 different 100+ page specs (some exceeding 500 pages) needed to properly parse excel files. An incomplete list (which you can find by searching for the keyword):

MS-XLS, MS-CFB, MS-ODRAW, MS-ODRAWXML, MS-OGRAPH, MS-OFFCRYPTO, MS-RMPR, MS-OVBA, MS-OSHARED, MS-CTXLS, ECMA-376, MS-OLEDS, MS-OLEPS, MS-DTYP, MS-XLSB, MS-OFORMS, MS-VBAL, MS-OAUT, MS-DCOM, MS-XLSX

But if you are looking to contribute to an open source project, consider some of these reasonably active projects (I'm sure I am missing some really good libraries here, so don't consider this a comprehensive list):

https://github.com/tealeg/xlsx (go, xlsx read)

https://poi.apache.org/ (java)

https://github.com/Niggler/js-xlsx (javascript, xlsx read)

https://github.com/Niggler/js-xls (javascript, xls read)

https://phpexcel.codeplex.com/ (php)

http://www.python-excel.org/ (python, based on openoffice )

https://github.com/randym/axlsx (ruby, xlsx write)

https://github.com/seamusabshere/remote_table (ruby, xls/xlsx read)

Anything I've ever needed any of the OpenOffice/LibreOffice tools to do have worked fine. I don't think it does anyone service to suggest to projects which keep the proprietary Microsoft Office suite alive.

Joel Spolsky (member of the office development team) attributed their success to this:

> Microsoft Office ended up beating its competitors because it was able to read and write to file formats other than its own. For example, Excel was able to read files from Lotus, and was able to save the file without losing any information. The important lesson in order to gain market share was that in order for new users to try Microsoft Excel, they had to be able to work with the files their coworkers were creating.

> Lesson: You don’t want your customers to feel locked into using your software. It helps with sales cycles because customers know that they can switch away if they so choose. While customers will be willing to try your product, they think two steps ahead. They want to ensure they can easily migrate to a new system.

http://www.bothsidesofthetable.com/2011/05/07/spolsky-on-sof...

Excel is 28 years old and for its entire lifetime it has been one of the focal points of Microsoft development. It is also the center of one of the biggest cottage industries in the software world + servers as an SDK for probably hundreds of thousands of in-house developers.

I think such an effort is just an order of magnitude smaller than replacing Windows on desktop systems. And that's saying a lot about Excel :)

That is one full time developer for a year, not counting any project management etc. If we say that so far, there has been spend 1000 man years on Open/Libre Office (I'm just making that number up, it has been under development for decades and probably with more than 20 devs working on it, so it's probably more, but let's say 1000, and not even count all the project management / user feedback / documentation work etc); then still you're saying that bringing Libre Office to the level of MS Office is within 0.1 resp 1 % of completion. I hope I don't need to argue how that is prima facie nonsense.

http://www.joelonsoftware.com/items/2008/02/19.html

(I realise I'm not answering you. I'm bad at estimates)

More than that, because you have to replicate a lot of Excel's idiosyncracies in order to get compatibility. That's a very challenging task.

When Microsoft was putting Lotus 1-2-3 compatibility into Excel back in the late 1980s/early 1990s, this was a much smaller task than replicating Excel would be today. Lotus 1-2-3 was a much less complex product. Most people were upgrading from 1-2-3 v2.x to Excel, which meant that Excel really only had to be compatible with a product that was originally hand-written in x86 assembly.

I wonder if there will ever be a shift to a future incarnation of hypercard, or some other mainstream oriented 'programming' platform that would displace Excel.

Word, Excel, Powerpoint all have viable alternatives in LibreOffice. I also really like OneNote as well. The biggest reason I don't use the MS products more, is that I happen to run Mac, Windows and Linux regularly on my different computers. More of my development work is now targeting Linux than Windows, which leaves me less tied to it.

I'd also like to see the Office org split out, and let free to target non-MS platforms more.. would love a better Office solution for Android. And, I'm sure iPad users would appreciate it too.

Anyone engaging in any sort of historical research?

Replace "Word" with "HTML". Now is it still the abominable dichotomy the OP is claiming?

Although I think Word sucks in many ways, letting users combine style sheet and local formatting doesn't seem like the Original Sin from which all evil flowed.

They didn't use a binary on-disk format by accident, nor was it a mistake. The folks who wrote word knew that what users would want to open and save files as fast as possible, on hardware thats weak and tiny by today's standards. Going for a format that resulted in the smallest possible files and the fastest possible reads and writes makes sense in those conditions.

Say you had a large document of 600 KB size. Floppy drives wrote at 45 KB/second. Imagine waiting 13 seconds for your file to save out. You might save less often -- which means that you ran a correspondingly higher risk of losing data.

The .DOC file is a binary format so that it could contain document "sections," with pointers between the sections. This is what made "Fast Save" possible. If you only made a small change to an enormous document, Word would simply append the changes, and then change the pointers in the rest of the document.

Instead of waiting 13 seconds, you'd get the save in under a second.

And note the 'IMO' bit: Why on earth do people think that because they don't like a piece of software, they want to force all the other happy users to stop using it. It's just selfishness and self-importance. The world won't just use the pieces of software that you want, you are going to have to live with other people's choices.

Choice- singular. The real problem with Word, as highlighted in OP's post, is its longstanding market dominance & attitude of outright hostility towards interoperability.

"The reason I want Word to die is that until it does, it is unavoidable. I do not write novels using Microsoft Word. I use a variety of other tools, from Scrivener (a program designed for managing the structure and editing of large compound documents, which works in a manner analogous to a programmer's integrated development environment if Word were a basic text editor) to classic text editors such as Vim. But somehow, the major publishers have been browbeaten into believing that Word is the sine qua non of document production systems. They have warped and corrupted their production workflow into using Microsoft Word .doc files as their raw substrate, even though this is a file format ill-suited for editorial or typesetting chores. And they expect me to integrate myself into a Word-centric workflow, even though it's an inappropriate, damaging, and laborious tool for the job. It is, quite simply, unavoidable. And worse, by its very prominence, we become blind to the possibility that our tools for document creation could be improved. It has held us back for nearly 25 years already; I hope we will find something better to take its place soon."

--TFA

So, you can disagree, but the "why" of your first comment is pretty directly addressed by the author.

I do feel that this is one of those cases where you should be able to write in the way you feel best for you, and if the publisher insist on having the final document in a word format, it should be straightforward to convert your chosen representation to theirs. It's hard to say without knowing precisely how the publisher expects the file to be formatted, but if it's fairly straightforward, there are libraries that will write to doc files for you, or there are open formats that MS Word already knows how to convert into .doc files that you could target instead.