Jeff Duntemann's Contrapositive Diary Rotating Header Image

How Necessary Is Windows? Part 4: Format Lock-In

Computing certainly isn’t about operating systems. Nor is it really about apps. It’s about files. Data is what we create, modify, store, and distribute in electronic form, and the ways that our data is stored give shape to almost everything else we do in computing. Being able to move from one platform to another thus depends almost completely on whether or not we can bring our files with us.

I’ve been working in front of a personal computer on an almost daily basis since May of 1979, and over the past thirty years I’ve accumulated thousands of made-by-hand files. Much of that is text, and I’ve had almost complete success bringing document files forward down the years, bouncing from one word processor to another by using various format-conversion tools. SF stories I wrote in CP/M WordStar in 1980 have passed through WordPerfect for DOS and several major releases of Microsoft Word and still live on my writing projects thumb drive. I keep a commercial Windows utility called Quick View Plus on hand to extract text from extinct file formats when necessary, which has been pretty rarely in recent years. Still, it’s there if I need it.

It’s a lot tougher once you get away from text. There are no conversion utilities for Adobe InDesign or Microsoft Visio, and as best I know nothing will import files created on either app. This is probably also the case for QuickBooks, and probably a great many more major applications that I’ve simply had no need for and no experience with.

I don’t think it’s an exaggeration to say that this kind of file format lock-in is the only thing keeping some of these companies profitable, or for some of them simply in business. Computing is mature in terms of the basic mechanisms we use to manipulate data: text editing, page layout, spreadsheeting, presentation, raster drawing and vector drafting, image processing, and database query and display. Small points may be patentable, but the fundamental machinery is now older than any surviving patents. Building an app that could load and edit an InDesign layout file would take some work but no genius, and if done would be a major competitor to InDesign, not only in new projects but (crucially!) in existing projects as well. Adobe guards its file formats with its life because its file formats are its life.

Alas, my files are my life too. And when Adobe’s IP rights bump up against my IP rights, who wins? Adobe, of course. Hence my hatred of activation: I don’t use newer versions of InDesign because Adobe can turn them off remotely and basically hold my work hostage if they choose to, perhaps because they’re too stupid to tell a RAID from a separate machine, or because they’re hungry enough to want to force me to pay for an upgrade that may have no new features that I need…and maybe no new features at all.

I have no general Stallmanesque animus against commercial software. I have a fortune in boxed apps on the shelf, and have never minded paying for them, even when they were upgrades. However, upgrading must be my choice, and migration of software to newer machines as time passes must not require new licenses.

Even commercial software that doesn’t require activation often demands a service and a tray icon, constantly popping up notifiers trying to upsell me to something I neither want nor need. In a mature market, there’s less demand for upgrades, and people can be happy using software for a long, long time. I can understand the vendors’ perspective and their need to be selling all the time to stay alive in a mature market. I think they should recognize my right to find it annoying and turn to software that doesn’t yammer so much and waste my cycles.

The bottom line here is that some apps are difficult to move away from; for me, the two killers are InDesign and Visio. Yours may be different, but I think most people who do creative work at the keyboard have a few. The difficulty lies entirely in proprietary file formats, and leads me to the infuriating conclusion that Windows is necessary only to allow me access to my own files.

The good news (for small values of “good”) is that there are tricks to be played. More in the next installment.

6 Comments

  1. Pablo says:

    I’ve never thought of activation as something that can be used to turn off apps remotely after they’re activated, in fact, it’s trivial to have the affected machine completely air gapped.

    The only thing I can say is that recent adobe products seem to be more liberal about importing from/exporting to more generic formats. That alone may be worthwhile.

  2. Jim Tubman says:

    An alternative to InDesign, which allows you to keep all your files in plain text (ASCII or UTF-8) would be Donald Knuth’s TeX family of software. You can do some very fine work with it; see

    http://www.tug.org/texshowcase/

    The software is free and open source, mature and robust. It is no longer limited to the Computer Modern fonts; modern versions handle OpenType fonts as well as the inclusion of graphics in many formats. The output format is now PDF by default.

    The downsides are learning curve and lack of WYSIWYG. I have called TeX “the ham radio of typesetting.” You have to learn a lot to make it do more than basic (but beautiful) math and physics papers. The software costs nothing, but realistically one needs to buy a few books about TeX/LaTeX to learn how to make it sing, and although they cost less than InDesign, of course they aren’t free.

    Lack of WYSIWYG is not a total killer. On Mac OS X, there is an environment called TeXShop that keeps two windows open: one for editing the source text, and one for previewing the PDF output. There is a similar program for Unix/Linux called TeXWorks, which I have not yet tried. But either way it does not allow direct manipulation of the formatted text.

    A couple of years ago, after I promised to help a friend typeset a parish history she had written, I took a look at TeX again, 15 years since my last brush with it. We were able to get some very good results with it.

    That said, InDesign is an excellent program (an iPhone to TeX’s ham radio, perhaps) and seems to be the standard for publishing. Giving it up would not be painless, but it is possible.

    1. There’s a conceptual gulf between TeX and InDesign that’s hard to explain: WYSIWYG vs. WYSIWYM. The TeX community believes that complete separation of content from presentation is possible, and I do not. How content is presented can either help or hinder understanding. Column width matters, fonts matter, page layout matters–and unlike HTML on a display, when your presentation is a printed page you can’t jigger things to make it easier on the eyes. The hard truth is WYPIWYGAIDWBBR. (What You Print Is What You Get And It Damned Well Better Be Readable.)

      That said, the LyX product has a lot of merit and I played with it a couple of years back, but I don’t think I can accept the underlying philosophy enough to use it comfortably. This may be because I am a publisher, and an old-school, page-centric one at that. I love layout for its beauty as well as its utility, and for that, WYSIWYG has to be the foundational approach.

      1. Jim Tubman says:

        [This spirit of this reply is “friendly conversation over beer,” not “religious/fanboy.”]

        I did a screen capture of TeXShop on Mac OS X; you can see it at http://members.shaw.ca/jtubman/texshopdump.jpg

        As you can see, you can indeed get a very good idea of what the final output will look like. TeXworks for Linux is not there yet but the goal is it to be basically the same as TeXShop, but for Linux/*BSD etc. We shall see.

        Based on what I have seen and read (and I’ve been a member of the TeX Users Group for a couple of years now), the TeX community isn’t really as hung up on the idea of complete separation of content from presentation as they once were. Once graphics start to get incorporated into a document, the whole “logical structure, not appearance” thing (which is really a LaTeX thing, not a pure TeX thing) starts to crumble. But it gets you 85% of the way there, before you have to start inserting manual line breaks and such.

        These are some examples of good layout that can be done with TeX:

        http://www.tug.org/texshowcase/leaflet.pdf
        http://www.tug.org/texshowcase/0309Newsletter.pdf
        http://www.tug.org/texshowcase/pp.pdf

        All that said, an open-source equivalent of InDesign with a fully documented file format would be a great thing for many people, including you. Scribus doesn’t seem to be quite there yet.

        This series of diary entries is very enjoyable and thought-provoking. I would be interested in your thoughts on what kind of social/economic arrangements might make it possible to produce complex open-source software packages (like an InDesign equivalent) with long-term support. Canonical has shown us one possibility with Ubuntu. There could be others.

        1. Bob Halloran says:

          At the Large Company I work for, we use the JBoss J2EE appserver. There are two versions available: the “community” build from jboss.org, with support from message-boards, etc., and the “Enterprise” build from jboss.com/Red Hat, that bought the company a couple of years back. The commercial version integrates various pieces such as message queueing, the Hibernate and Seam frameworks, etc, and provides 24/7 support. They will not accept trouble tickets on the community build. Obviously a number of the developers are on the Red Hat payroll. This dual-track mechanism would seem to lend itself to the sort of large projects you describe; get the “hobbyist” version for free with the understanding your support is limited and you may have some work to fit it into your environment, and a “professional” version with better support and integration.

        2. There are indeed others; the one I’m most familiar with is Crossover Linux from Codeweavers. It’s a $40 product that basically gives you a generic installer for Windows apps under Wine, plus support for installing a list of popular Windows apps under Crossover. There’s a $70 version with more features and support after app install, renewable after 1 year as a $35 subscription.

          There’s feedback to the open-source product: Code improvements plus a certain portion of each sale go back to the Wine community for supporting ongoing development. Wine is all in the emulation, and doesn’t have a noob-level installer. Crossover does. They seem to be doing well, and when I feel better I’ll continue the series with an entry on Wine and Crossover.

Leave a Reply to Jeff Duntemann Cancel reply

Your email address will not be published. Required fields are marked *