Why would anyone want a word processor?

Tim Bray “describes his woes”:http://www.tbray.org/ongoing/When/200x/2005/11/12/Resume-Blues working with his résumé, which was being maintained in a MS Word document:

It’s got some columns-and-anchors magic where the dates are out in the left margin and are supposed to stay attached to the jobs and publications and so on that they belong to. Only they get out of alignment, and then when I try to move them by hand, things go weird, the grid is wrong and when I hold down the magic free-movement key they start oozing out of the margin. Then the hanging-indent in the little bulleted lists stopped working, so I was having to nudge the continuation lines over. I guess at some level it’s creditable that this misshapen outgrowth still works at all, but anyhow my patience ran out.

He then describes how he converted the document to HTML, and asks:

So, I understand why we still need spreadsheets and presentation packages, but assuming you had a Web editor with a good change tracker, why would anyone want a word processor any more?

That question is one I would like answered as well. I personally edit everything that has anything to do with text in a web editor (or should I say “text editor”?), and I keep it in a “Subversion”:http://subversion.tigris.org/ repository to keep track of changes.
Adding to that, I also edit many of my spreadsheet-like documents in a web editor. Or should I say: I edit most of the tabular documents I maintain in a web editor as HTML. And, anything vaguely presentation-like is also done in HTML.
_So, why do people need office packages at all?_ Granted, I don’t expect the layman to tinker with HTML manually, nor do I expect them to understand a revision control system like SVN. But: Why are there no good web editors for these people, and why don’t people use online office suites? Even the simple spreadsheet is easily solved with a little JavaScript.

Previous Post


  1. Personally, the only reason why I’m using OpenOffice rather than HTML for my master’s thesis is that HTML has no good way of a) dynamically keeping track of footnotes, and b) displaying footnotes in a page context. HTML is not great for paged media, though you can work around a lot of the problems with JS and CSS.

  2. Eira touched upon possibly the most crucial aspect: paged media. Most people want a fine control over what the pages look like, and, if needed, want to change the layout on an ad hoc basis.
    Not only is HTML/CSS limited in its paged media possibilities, current browsers also cannot produce a decent print output, and Prince is too expensive. I ran into that problem after I had written my CV in HTML and tried to create a PDF of it: the quality of the print output sucked and I had to rewrite and reformat it in OpenOffice.
    Personally I considered writing my master’s thesis in HTML, but switched to LaTeX, mostly due to the bibliographic tools, internal referencing (e.g figure numbers) and especially the high-quality print output.

  3. Jud

     /  2005-12-20

    SubEthaEdit is a rather interesting-looking collaborative text editor. Haven’t used it myself. Makes use of Bonjour/Rendezvous/Zeroconf (SVN uses WebDAV, another collaboration technology, so your mention of SVN reminded me about SubEthaEdit).

    Seems to be aimed at collaborative programming more than writing tasks such as footnoting thesis pages, however.

    Mark’s remark about LaTeX is reminiscent of others I’ve read, many of them from people trying to write theses and finding that MS Word or HTML just wasn’t working for them. Guess that Don Knuth is a pretty smart guy. 😉