Advogato blog for davidr

10 May 2003

Sat, 10 May 2003 23:23:51 GMT

Well, it finally happenned! The server has died yet again and given it's somewhat strange departure (dying and being reborn a number of times overnight before it's final curtain call) it might be fatal this time. Time to source a new server and get the old one replaced! Of course startig such an undertaking at 5pm on a Saturday isn't ever going to work, and as I'll be away for much of next week it'll be even more stressful than it might otherwise be :(

I'm hopeful that tomorrow after a swift kick from a size 12 it'll rejoin civilised society and give ma the time to ease the transition, but my luck isn't normally that good so no money will be riding on that bet!

Some interesting developments wlesewhere in life though I still find myself battling a cross-compile of gcc at present. That battle has a few rounds to play out yet, but until the server issue is resolved we've agreed a truce :)

4 May 2003

Sun, 4 May 2003 09:26:38 GMT

I don't think that many folks here will have seen this, but while the english in the post isn't perfect it's well worth reading. Frans has been a friend for a while and I was very sad to read this :(

http://www.xentronix.com/module.php?mod=node&id=276

He's very open and honest and I wonder how many people who frequent this site can't see at least some of their own behaviour in what was written?

7 Apr 2003

Mon, 7 Apr 2003 11:40:45 GMT

April already! Wow, time really does fly :)

I've been following the developments in the BeOS world over the last few weeks (http://www.beosjournal.org is a useful site if you get over the horrible look) and I've got mixed feelings.

I've written here before about OpenBeOS and why I left that project, but being honest looking at the website and browsing the recent newsletters they've posted I'm glad I did. The in-jokes, constant Palm baiting and general feeling of standing still are overwhelming. Recent newsletters seem like they're being written just to fill space! Sorry, but the one thing that the project needs is to be taken seriously, and recently they've been doing more and more to move away from that!

I appreciate that the BeOS community wants something to look forward to, but I really think that the love affair they have with OBOS and the ignoring of it's faults isn't healthy for the community or OBOS.

Zeta looks much more interesting and may well be what the community have been waiting for. They've played the game close to their chest with few details being released and that has attracted some critisism, but overall I think they've done it the right way. Results look promising and the pre-release "buzz" is that it's good. Good enough to make an impact? We'll have to wait and see but let's hope so for Yellow Tab.

20 Mar 2003

Thu, 20 Mar 2003 23:39:31 GMT

Hmm, forgot to mention that I actually did some work on my idea for a central place for driver information, though it's not yet finished! Where am I supposed to find time for all the things I want to do? Someone invent a way to give me 36 hours per day and I might make more progress :) Anyone who's interested can find it here

http://www.camodi.org

If people want to help out then that'd be cool so they should email me direct at david at jetnet dot co dot uk.

20 Mar 2003

Thu, 20 Mar 2003 23:31:33 GMT

Wow, has it really been that long since I posted? Guess it has...

Life has been through it's usual cycle of ups and down since my last posting, though generally on an upward slant. I'm still developing and writing code though not in as public an arena as I have been for the last few years. My decision to get involved in what is essentially a closed source project will surprise some folks I'm sure so I'll try to outline why :)

Open source projects need only one thing to function well - a large body of interested people actively developing the code. That's the conclusion that I've come to after several years of being involved in various oss projects. The lack of a large body of people working will result in one or two doing the majority of the work, with little or no review of the code and the resulting bugfest will be immense in proportions! Bugs are only one of the problems though. OSS projects by their definition don't tell people what to work on so people scratch their own itch. With alarge body of developers this isn't a problem as invariably there will people scratching all the itches, but as the body of developers shrinks so to does the number of itches being scratched. The recent spate of postings about XF86 seem to echo my feeling quite well.

In the situation where you have a few developers working on a project the other problem that gets generated is all the "hangers on" who feel that they can comment on all aspects of the project. done correctly and in a constructive way this can be very helpful, but more often than not the comments are along the lines of "why don't you do this..." with no suggestions or helpful comments. How often have you been working on a prohject and read such emails? Let's be honest, they're not very helpful or encouraging are they?

The combination of these problems has led me to where I'm currently devoting my time. The source may be closed and the development team is small, but we are following the best OSS principals in how we approach the work - CVS, mailing lists, IRC and peer review of commits.

Don't get me wrong, I think OSS is a great motivator and in many cases produces better software than would otherwise result, but I don't think it's right for all projects. The existance of openly available code is wonderful and empowers much development that wouldn't otherwise be possible.

23 Dec 2002

Mon, 23 Dec 2002 11:52:08 GMT

It's that time of year again. Christmas is upon us and soon it'll be the New Year with all the changes that brings. It's also traditional to look both back and forward at this time of year and as I've been doing that here are my thoughts :)

Personally this year has seen highs and lows with a life altering experience being the main incident that stands out. It really brough home how valuable friends are and how much they can help when the days look blackest. A big vote of thanks goes out to all those in the open source community who were there for me when I was in need (they know who they are!) :)

In terms of my involvement with open source the year has been mixed. I've been quite depressed by the way that anumber of projects I was/am involved in have changed over the last 12 months - mainly for the worse. One of the things about the oss community that I always liked was the way that people were allowed to express opinions and a level of respect was given by the rest of the community. This tended to lead to a nice collaborative environment in which good work was possible and even encouraged. Over the last 12 months I've seen this ideal disappear and be replaced by an attitude towards people and ideas that borders on the aggressive. Left unchecked this change is truly destructive and could well be the start of the end of many projects and their associated communities.

My day job requires that I have the ability to work with a diverse group of people in a challenging environment, often with situations that can rapidly escalate into problems. It's amazing how often a misspoken word or a percieved slight can cause communications problems. Email compounds this vastly and if I had a dollar for every time I've seen this sort of problem this year I would be a very rich person indeed. Quoting people's emails out of context or missing relevant sections ot alter the meaning is a common tactic used by some on mailing lists. Such tactics are ultimately self defeating and cause untold harm to the community. Given that the communities in general feel that they are mature enough to control and police themselves (and some have reacted violently to suggestions that more oversight would be a good thing) then these childish tactics don't fit with this vision - so why do people continue to use them? I suppose if I could answer that question I wouldn't be sitting here about to go to work :)

As for the coming year, I'd like to think that we (the internet community) would start to exert our power a little more. We are a large group of people with a diverse set of interests and desires. This gives us untold power. What do we do with it? Very little. There are many places where people complain about things, but very few people do anything about it. Companies exist by selling products to make money. If the community got together and decided that they were going to send a message by boycotting a product this would be a clear signal that couldn't be ignored. Email campaigns are all well and good but they are easily ignored. Slashdot is a great site for geeks, but how many CEO's care about the contents of it's stories? Maybe 2003 will be the year we finally get a clue and start to exert our muscle. Obvious targets include the copy-protection of CD's.

Maybe if everyone who owned a computer stopped buying CD's for a month the companies would get the message? Maybe if we stopped buying just the CD's with the protection we'd send a clearer message? I think so and if anyone has a link to a web site that gives details of which CD's are so protceted I'd like to know :) January 2003 would be a good month to start with!

Well, that's some of my random thoughts about the year past and the year to come. I'm involved in some projects that will hopefully reach releases in the first 6 months of the year and so I forsee me being busy for quite a while to come!

Happy coding :)

7 Sep 2002

Sat, 7 Sep 2002 10:09:32 GMT

Had made good progress on my Win32 project but now, alas, I've been called upon to go to work :( Hopefully I'll complete the win32 stuff so I can get back to doing some Apache development upon my return next week.

Also, I've almost done with repopulating the photot gallery so it should be ready to be moved over to the live sites before too long, which will be another lon term project completed :)

Well, bag packed, time to go!

Enjoy the weekend.

4 Sep 2002

Wed, 4 Sep 2002 11:48:19 GMT

Well my last diary entry actually provoked a response! Wow people do read these things :)

Well, in a moment of weakness I bought Warcraft 3 and have been playing it for a day or so now. It's good but not as good as the C&C/Red Alert games that are the best I've played. The interface is OK though I don't think the game quite lives up to it's hype.

I'm working on my revised photo gallery code and it's coming along though I'm having trouble deciding how I want it to look and then trying to get the html to display something like I want! Still, it's getting there and then I have the harder task of merging together 3 older sets of databases that drive the various web photo galleries I have.

This week sees the start of 3 weeks of standby at work so I just sit at home and wait for the phone to ring. Could be an opportunity to get some serious coding done though :) Now if I can just get the interest level up...

1 Sep 2002

Sun, 1 Sep 2002 12:07:14 GMT

Can't figure out whether I should write this as an article, so (due to the large yellow stripe down my back) I'll write it here and it'll get no coverage, but I'll feel better :)

What am I talking about? I'm talking about everyone concerned with "alternative" OS's (by that I mean anything that isn't Windows) working together and boycotting certain hardware manafacturers. Why? Well, there are a lot of hardware companies that publish enough details of their hardware that people with enough time and desire can write drivers for any operating system. There are however a lot more companies that don't publish any details making drivers almost impossible to write in some cases.

These companies only seem to understand sales and have ignored efforts, in some cases concentrated efforts, and requests to have the details made available. Whilst this is annoying it's actually counter productive but at the moment it's hard to figure out ahead of time whether a company is "alternative OS (AltOS) friendly" or not. It's time we fixed that and published lists of companies that are willing to work with the AltOS community. Of course we should also publish lists of companies that are NOT AltOS friendly - boycotting their products totally as this seems to be the only form of signal these companies will understand.

The commonly supplied arguments against releasing enough specs to write a driver never seem to hold water. 1) It's propriety and we don't want to give our competitors the information.
Surely these companies aren't niave enough to think that as soon as their products are available their competitors don't do some reverse engineering on them anyway? 2) It costs too much.
As the information must be avilable to these companies own engineers why not simply publish it on the web? It doesn't have to be pretty or well organised(though it's better if it is) as long as there is sufficient detail. I mean how much does FTP cost?

Those are the most common arguments I've seen and as I pointed out I don't think either really hold water. Given the way the industry is moving I really don't understand why companies are so resistant to having their products useful on as many computers as possible. The windows market is shrinking and will continue to shrink. Despite the Linux fanatics shouting there isn't a single OS replacing it nor will there ever be (IMHO).

So, here is my first list of companies I won't be buying anything else from and would urge others to follow my example.

- Logitech - Sony (memory stick port and jog dial thingy on their laptops)

I'm sure there are others that I meant to add and have forgotten!

25 Aug 2002

Sun, 25 Aug 2002 22:32:50 GMT

Well, another trip done and I'm home again!

Met up with Rene in Dallas and we went to the JFK museum which was cool and more interesting that I thought it would be. Quite amazing to see just how much the site looks like it does on all the pictures of the time. Expected it to have been altered, but they've kept it the same.

Openbeos's net team seem to be changing their direction for the worse. http://www.freelists.org/archives/openbeosnetteam/08-2002/fullthread41.html While I was team led I integrated a lot of the "lower" level code into the kernel to get the best performance from it and keep it as simple as possible. With the plan they're outlining and planning to use things will be worse and the performance issues that beos networking have always had are likely to continue. Every time I read this sort of thing it makes me glad I left the project when I did.

I'm trying not to be too negative (in the bitter and twisted way that so many people get when leaving project) but the progress being made is all superficial and "edge" stuff that really doesn't move the project on, in my opinion. Ah well. It could have been a good project, but it doesn't look like it'll ever get anywhere now.

My driver continues to progress with ISO transfers now being run. I'm still fighting with trying to figure out how to size/create/use the ISO buffers correctly under the beos model as the buffers I pass in are always returned as NULL's. Oh and I have kernel crash on exit that I think is due to my driver :) Still, a lot of progress in a short time.

Time for some well deserved sleep now!