Blog This!   Lee Geistlinger's Web Log

Blog Home
Blog Archives
LittleGhost Home

E-mail: geistlinger AT gmail.com

Everyone loves lists Books, Movies and so on

My Online Aggregator

Open-Source Silliness

Tim Bray points to what it a very odd ZDNet blogger entry by Dana Blankenhorn suggesting - well, saying - that open-source applications are overwhelmingly just followers, not innovators. (And so what's the use of it?)

While there is a certain degree of truth to this - Open Office is replicating MS Office, the GIMP trying to give the world a free Photoshop and so on, this isn't the whole story. Because you could say the same thing for closed source software.

• MS Office was just a WordPerfect and Lotus 1-2-3 clone with some other tools.


• Aldus Freehand tried (tried...) to duplicate Adobe Illustrator.


• Aldus PageMaker was a QuarkXpress copy, and now Adobe has their InDesign (I think that's what it's called).


• On the Web scripting language side, I don't know who came first, but I guess the first one - to Blankenhorn - is the only innovator, the rest mere clones, be they OSS or CSS. So only one the following (abbreviated list) is an innovator, the rest imitators: PHP, ASP, ColdFusion, JSP, Lasso.

I really don't see a lot of breakthrough apps on either side; sometimes the first innovator remains the best (Apache) in many respects; in some cases, the innovator falls by the wayside as others jump on this band wagon (Netscape eclipsed by IE, later by Firefox and Safari; Radio blogs bested by SixApart and Blogger). There are different types of innovation: breakthrough and incremental, for example.

And let's take a look at a handful of OSS projects that were truly innovative and still remain the leaders:

• Apache - OK, I think this began as NCSA Server, but this was before the Web itself existed. Apache currently has approximately a 70% share of the server market, a figure that has remained pretty consistent (with an overall uptick trend) for the past five years. Apache runs the Web. Even Microsoft, with its large installed base and millions in advertising yearly, is second with roughly 20%.


• Perl - Before Perl, there were handfuls of shell and C scripts. Perl still massages the data that is presented on the Web. Python - and now Ruby - have made some inroads, but Perl is still out there and running strongest. Virtually every Unix server - and many MS servers - have Perl on them; few have Python or Ruby. First of its kind; still the most popular.


• Javascript- Developed by Netscape, this language (whose name was changed from LiveScript to cash in on the growing popularity of the then-new compiled language, Java [NOTE: JavaScript != Java except for some similarities of syntax]) is now the engine (along with CSS) that is AJAX: Part of what some call Web 2.0. VBScript is the only competitor, and only really used with ASP.


• CSS (Cascading Style Sheets) - Interestingly, I have no idea where this came from; I think it might be a W3C implementation. I just don't know. All I do know is that it's not a shrinkwrapped product. It's a spec, more than an app.

Also, Blankenhorn missed the mark - to me - by suggesting that apps are only innovative if they are totally new (for example, he points to TiVo, which is a good example of such; another [more retail] would be Netflix).

But sometimes the way an app/software is innovative is in the way you can use it, or how you can build it. JavaScript, for example, is a weak language. But you put it in a browser and can do things client-side: No server hit. Whoo-hoo! This made it powerful if it were the first or 1,000th incarnation of this type of language.

And do you every wonder why there are so many (according to surveys) VB coders vs, for example, C++ coders? Two reasons: 1) VB is built to be easy to use; 2) MS's Visual Studio is an amazing work environment. Easy - but powerful - language with a great IDE. Duh. So we gots lots of VB apps.

If you applied Blankenhorn's statements to cars instead of software, well, hell, the Toyota Prius, Audi Quattro and Hummer H2 are all just efforts to replicate the Model T (or perhaps Daimler's and Benz's first horseless carriage...).

Blankenhorn also misses one very important point, one he almost captured...but missed.

Projects like Mozilla and Openoffice are all about offering free replacements to proprietary monopolies. Databases like mySQL are still working on "innovations" proprietary products had years ago.

This is true in Linux as well and it's not necessarily a bad thing.

See - he almost had it: One of the truly innovative aspects of open-source software is that it is about offering free replacements to proprietary products.

View this as a good thing or bad (Blankenhorn agrees, at least for Linux, this is a good thing), it's definitely new.

And innovative, Ja?