Flutterby™! : 16 categories

Next unread comment / Catchup all unread comments User Account Info | Logout | XML/Pilot/etc versions | Long version (with comments) | Weblog archives | Site Map | | Browse Topics

16 categories

2001-02-24 19:45:13+00 by Dan Lyke 4 comments

Dave Winer recently posted a list of 16 categories where commercial software beats open source. I emailed him a note, he asked if I'd post it so he could comment in public, so here's my notes on more on the 16 categories. Update: Dave has responded, I think we've many of the same concerns.

[ related topics: Free Software Dave Winer ]

comments in ascending chronological order (reverse):

#Comment made: 2001-02-25 11:37:59+00 by: crasch [edit history]

From my limited reading on the spread of innovation, you tend to get an S shaped curve when you graph the number of adopters of a new technology/social practice as a function of time. (See Everett Roger's Diffusion of Innovation).

I think that open source software is still close to the base of the steep part of the curve. I expect that open source technologies will come to dominate most of the other categories eventually. For example, PostgreSQL has made great strides in features and stability in the past two years. Now that GreatBridge has begun to pump $35 M into PostgreSQL development, I expect that in the next year or two, it will become technically equivalent to Oracle for all but the most demanding applications. As a result, I expect that it will make increasing inroads on Oracle's territory, as more companies have positive experiences with it, and as a rich support community grows up around it.

It also takes time to erode established market share--it took Microsoft about 15 years to whittle Apple's share of the PC market from > 50 % to < 5 %. What will Winer's list look like in another 5 years? I don't know, but I expect quite of few more "best of breed" apps will have switched to the open source side of the ledger.

#Comment made: 2002-02-21 05:31:11+00 by: petronius

A question about open-source is: who is developing this stuff? For a graphics program, we want fast and subtle image manipulation. However, we also want formats and tools compatible with the printing industry, like CMYK color systems and easy use with Postscript printers. PhotoShop is huge and expensive, but works beautifully with the needs and desires of the graphics industry. Are the open-source people getting the programmers to work with the artists to get these tools created? Commercial houses will identify a need, and hire both types of specialists to do the job.

#Comment made: 2001-02-26 18:12:25+00 by: Dan Lyke [edit history]

Re: PostgreSQL. It's only really been viable since release 7, and for many of the places where it could replace MySQL only 7.1 gets close, and even there there's a learning curve. Even though I've thrown my lot in with PostgreSQL[Wiki] I don't necessarily want it to "win", because I think the competition from the MySQL[Wiki] folks has been extremely valuable.

On the "who's developing this stuff?" question: There's a reason that there's development on deep color in GIMP, and color matching is kind of lax. The film industry has one paint program, Amazon paint, and it's not everyone's favorite. In film you need deep color to account for the increased contrast range of film over paper, and color matching is something you do in the monitors and film recorders. Furthermore, film industry users tend to be technically adept, even Amazon paint[Wiki] has programmable brushes and similar tweaks. Thus it's totally reasonable that people in that realm are hacking on GIMP[Wiki].

If you want to see the markets for open source, look for small numbers of highly specialized and skilled users, the kinds of niches filled by high dollar extremely vertical software.

#Comment made: 2002-02-21 05:31:11+00 by: baylink

I watched as the Apache people completely ignored (this is not a complaint) the development work we did with Frontier, adding an HTTP server, then Manila, without responding.

And the reason why Dave thinks that Apache *should* respond would... *be*?

It's not Apache's job to respond to Frontier.

Obviously, he hasn't been looking at *Zope*, which may well end up eating his lunch in another release or so. Or he's just trying to pull the same sort of crap the RIAA is this week: releasing an anti-Napster press release syaing CD sales are down 39%. No, CD sales are *up* 3.1%. CD-*Single* sales are down 39%.