Older blog entries for mattr (starting at number 25)


Though it took typing a nasty NVRAM patch into nvedit by hand, I finally got the unbootable G3 to boot NetBSD off of its hard drive. If I'd been able to boot into MacOS, it would have been a few clicks. But I do have a nice collection of Open Firmware links and a deeper understanding of those whacky Macs now... Open Firmware is pretty interesting, really. sigh Too much too learn.

Tried X real quick, and it's working at least. I never did get X to work right on my NetBSD/i386 box at work. I guess I'll have to get pkgsrc onto CD and install some goodies.

1 Jul 2002 (updated 1 Jul 2002 at 11:31 UTC) »
What a day...

July 1 was our first GOVIA developer's milestone. It shouldn't be July 1 already, but there it is anyway, so I'm coming off of a 12-hour hack session. On #govia, I cracked (as surely others have before) "This is the bizarre as opposed to either the cathedral or the bazaar." An observer on the channel, just before that, had said "sitting here _trying_ to read what you two are writing ;)"

I guess our (the two primary developers) communications are reasonably secure by virtue of the natural language er, something or other someone mentioned here recently... And trust there also an issue; have to address it later though.

It was a good 12-hour day... And now, off to sit in front of a different couple of computer for about eight hours!

Oh, and how interesting that I should become a 'Journeyer' today, all of a sudden, of all days. Er, thanks, but hm, this thing must be broken! There's one actual Journeyer cert, but the accumulated Apprentices add up to a Journeyer?

1 Jul 2002 (updated 1 Jul 2002 at 02:30 UTC) »
raph said:

The use of the word "trust" is interesting in this context. Is it the same trust I talk about with trust metrics and the like? In one way, yes. The real question is: who do you trust? No sane, thoughtful person would trust the likes of Microsoft and VeriSign. But I think people do trust their social network, and are right to do so. We need massive research to explore how to map social networks into the digital world. I often feel quite alone in thinking about this stuff; very few people seem to understand it. Oh well. I'll just keep pushing it patiently.

For what it's worth, this work you're doing on trust metrics and applying it to the free software community (those of us present here, anyway) is why I'm here. I wanted to see how it worked, which also, of course, helps make it work. Starting a diary to chronicle my minimal involvement with and laughable development of free software wasn't necessarily high on my list of things to do, but I see some real value in this research, just as I see tremendous value in the free software 'movement.'

Far too many folks these days aren't thoughtful people and have no problems trusting whatever corporate entities they're 'supposed' to trust, but if we (er, you? hehe) can develop and demonstrate an alternative that has some basis in our subjective experiences as individuals (rather than as various corporate account numbers), there may be hope. Time may be among the largest problems facing such an effort however. I know I haven't been able to go back and state my certification reasoning and can barely keep up with a diary simply because there isn't enough time to get it all done these days...

Er, that GOVIA developer's milestone set for tomorrow, for instance, should have kept me from even writing this... ;)

Sigh, more later...


Well, how about another section in my notes -- Certification Temptations! Just a spur of the moment thought from a quick perusal of the recent entries... Someone stating they don't deserve a certification, someone a couple of days ago wondering about the value of the certifications... A trend. I figure the more insight I try to provide in how I'm looking at the trust metric, the better. Now if I could just spend more time reading...

"Trust in me..."
Only suitably imagined as sung by Siouxsie on Through the Looking Glass.

The ol' 9 to 5

Though I'm in fact not 100% finished with the staff activity report web app, everybody's getting training on it today. I even received a couple of changes last night after 5pm, so I actually worked on it a bit at home. Reworked several things while I was at it; think it's a little less kludgy in places now. And my, how much easier it was to actually get some work done at home!

At any rate, no doubt I'll have some tweaking, fine-tuning, bug-fixing to do with it in the next couple of weeks, but I'm just relieved that it didn't blow up this morning. I wouldn't have enjoyed looking like an oaf, and I'm sure my manager (conducting the training) wouldn't have either. Furthermore, the iPaq it's running on is handling the load beautifully... Not that the load is that great, but it had been a concern, simply because I hadn't the foggiest idea of what hardware would really be necessary.

Now let's see if someone here develops some interest in how some freaky-looking kid in tech support did this with free software and a spare machine. The doc root index lists what I'm using, complete with links to the main sites and license information. Even threw the uptime in there... And the page title? LAMP Light @ [my company] :)

Hmm... there was more floating around in my head earlier, but I'm not sure where it went. More later, I guess.

Day job

Still making changes, even after starting the after-the-fact spec/requirement document for my pet project. We're still planning on training staff to use it on Friday though. Tomorrow will be a (perhaps long) day of wrapping up many details.

Then I need to rewrite it... It's become a huge mess with all the changes... Rather like I started with something that worked pretty well and have since been making personal preference 'patches' upon 'patches' to the point where it's scarcely the same thing at all.

Still, it's a joy to be doing something I like that much at work.


I really need to put some time in on some GOVIA things... It's been quite a month and I've been quite distracted. We're coming up on the first milestone and aiming at July 1. We can handle that. This weekend, some laundy, some hacking. I can really get things done on laundry days. Built in breaks for reflection when it's time to fold, then nice stretches of coding for the dryer... I mean waiting for the dryer. And thinking about socks.

Anyway, the submission process needs a bit more work and Gist is, although still working, hopelessly broken. Go figure. It was written when we were at, I don't know, version .2 or .3 maybe of the markup language, which we're now calling .6, so...


I was up entirely too late a couple of nights in a row trying to get the new G3 in the family to netboot a NetBSD kernel. I don't even have all the sets to install here at the moment (and I'm on dial-up at home :(), but it's driving me crazy that I can't even get it to boot. Somehow, though I used it a couple of days ago, the floppy drive won't let me insert a disk either, so I'm avoiding the machine entirely for a couple of days (or more, considering the GOVIA work I said I need to do...). I found some information in the port-macppc mailing list archives that suggested I might in fact be looking at a 15-minute wait (?!) for the thing to quit saying TFTP TIMEOUT and finally boot. Dunno, guess I'll find out eventually.

Also, the mailing lists I attempted to start a few days ago (well, no, also the mailing list that's a few months old) don't seem to be behaving. Something I mailed to one list early yesterday morning has yet to arrive at any of my subscribing addresses, and someone told me he attempted to subscribe to one of the new ones, but never got a response.

I've been more than happy with my web host in general, but having mailing lists that don't work doesn't do me any good. We'll see what support has to say. Maybe I did something goofy and broke it myself.

www.highmayhem.com is a pretty interesting site.

Christian Anarchy is a pretty interesting book. You can even read the whole thing online, though I bought a copy of the dead tree edition after I read it online.

Funny, I'm smelling coffee all of a sudden (there's none in the house)... Or am I dreaming, nodding off at work tomorrow already...

Lately I've been using a new alarm clock:

at [?:??] -f CDWAKEUP

Waking up to Clutch's Jam Room is interesting...

Hey hey now, what's that smell? Just like cornbread done too well.

24 Jun 2002 (updated 24 Jun 2002 at 01:20 UTC) »

It seems I've acquired yet another Mac, another beige G3. I had it here to try to 'help' it for a friend, but it has something interesting going on and will not boot after installing MacOS 9.2. Just won't do it. Not from the 9.2 CD-ROM, the 8.5 CD-ROM, a slimmed down 9.2-on-a-ZIP... Hangs and hangs and hangs some more, with or without extensions, and I can't even get to OF...

The owner said he'll just wait for his financial aid check in the fall and upgrade to a newer system (I intend to persuade him that his new system should be a GNU/Linux system, and am well on my way, I think), and said he has no use for this G3 since he doesn't want to put any money into it.

I never give up hope, of course... Maybe now it will feel the love of my LAN and consent to boot somehow... At which time it will get some repartitioning and some NetBSD. We'll see.

I've noticed several folks keeping track of who they certify and why. Sounds like a good idea, something that would be good as part of Advogato... ie certification includes "Why?" and insert that info with the list of who I've certified...

Would be beneficial to the metric, I think. As an unknown nobody and not-knowing anything, I'd check out the Recent Log and certify folks (and I have no idea why...will try to remember...), paying extra attention to fellow nobodies...But already since then my perspectives have been shifting...

And those looking at me might think, oh this guy's just certifying people to try to get certified himself and writing silly diary entries to waste bits... And I can't prove otherwise. Whatever's been distilled enough in my head to make it to my diary here is in no way indicative of my thoughts on open source software advocacy...Just informal rambling when I get a moment. But if my reasoning behind certification of so-and-so is exposed, perhaps others will get a better idea of my thinking.

Giving up hope in the trust metric doesn't seem like the best idea; we're learning as we go, no? Why not keep exploring? If some folks are stuck as Observers (I was for a while too), can you tell why? Is it because half of those here are more worried about what everyone else is writing here instead of what everyone else is actually doing?

Day job or
Adventures in Sneaky Open Source Advocacy

But only because I have no choice but to be sneaky.

I think the web-based weekly report project is nearing completion. So, of course, that means it's time for me to write a requirements document for it. Interesting, eh? Maybe I'm also sneaking in the 'bad' parts of open source software, projects begun with minimal direction, just a thought, an idea... But hey it works, to a point.

There was a need and an idea. We were given the go-ahead to see what we could do, without using any company resources. I started with what I knew, still learning daily as I went along. A couple of sitdowns, looking at the project with my boss, has refined the idea. Now that we think we have it, we'll "make it official," so to speak, with some documentation.

He gave me an example to go by, so I'll have to take another few glances at that. I hope to be able to emphasize the strengths of the open source stuff used in the solution, but I imagine a lot of it won't be within the scope of a requirements document. There is at least the one: "...without using any company resources." I assume that meant not to place additional demands on our web team, which we have avoided. My salary the past little while has paid for its development, but our department is well-staffed and the payoff will be significant, I think.

Where once several someones were kept busy cutting and pasting Word documents that had been emailed to them (looking back at my old ones, probably 40k average size) those same someones will now click a button. The database right now isn't even 25k (granted, only a handful of people have been using it, more or less as beta testers).

In addition, these serve to inform supervisors of problems employees might be having. We decided to go with an email notification system rather than including that in the report, so our solution facilitates rapid response to any issue an employee may have.

So now to document it. Well, not now, but you know.

Another mini-project to follow will involve keeping us under a maximum number of users at a particular web-based resource we're using in technical support.

In other semi-related news, another department (just barely another department, but it is) is also making use of open source software at the Day Job, putting together some web-based training material. The other Linux freak is over there with his Mandrake laptop, doing graphics with the GIMP, and writing the HTML in something other than vi(m), which he really oughta work on... -grin- A potential convert still harbors a desire to use ASP and a Windows server. Not sure how they'll handle that, what 'resources' are available. I told him the other day that PHP had passed ASP in the number of servers running it...

That's enough for now.

Day job

A bummer today... We do not support using our software with Linux servers, but we have a handful of folks doing it anyway. I've been pushing for us to change our stance on that support and have been in touch, informally, with some of those users. Seemed like a good idea to work up a brief survey, just to see what they were thinking... Whether they'll stick with Linux, what kind of problems they had, what type of configuration they're running with, etc...

The issue even came up yesterday in a meeting of bigger- wigs; specifically, there was some concern that we'll have users picking up the nice 'n cheap Wal-Mart Lindows machines and trying to use them with our software.

I finished a draft of the survey yesterday and had sent it on to my supervisor, who previously said it was a good idea. Unfortunately, when he took it another step up, it was completely shot down. Survey cannot leave the building.

I don't get it. Apparently our plan is to ignore anything new which is not MS-inspired. In the process, I feel like we make ourselves look silly when among the handful, we have users running other applications (Office, for instance) from Linux servers with no problems, but we won't even officially investigate the feasibility of running our software with such a setup. I've done it unofficially with no real problems, but that does not a changed support stance make.

paperlove.org & whonose

I created a couple of mailing lists yesterday at paperlove.org, one to discuss non-profit planning things, one for WhoNose development stuff... Don't expect either to be very active, but I'm hoping I might get a little input here and there. See this page for more on the lists, if you want.

tractor at Ozone Farm has been a big 'encourager,' so I thank him for that... I stumbled upon some really cool people when I was wondering where I could get my feet wet with Open Source Etc. :)

Other Stuff?

My Win2k box at work had to go down yesterday... A day of much virus scanning all over the building. The record will stand for now at 87 days, 20 hours+.

Fortunately, I had another machine I could continue to work on yesterday... Much productivity lost yesterday elsewhere, I'm sure...

Much has happened of late...

I really need to spend some time on my web site...or on the WhoNose scripts, really. But recent efforts have been directed towards GOVIA... We're approaching all-around functionality now, rather than just having a very crude search engine to see a dozen contributions.

The original XML format has changed a good bit; I'm not even sure the original crude search engine works as it should at this point. I'll have to re-fix it when we get other things more finalized.

Currently, we're focused (or are about to be focused...) on participant registration and getting the queue to work as it should.

Day Job
Difficult to get things approved by the right people and finalized... but the day job hacking project will probably be ready for use next week. Too much other internal training and meeting going on this week, plus I have a few things to change based on some up-the-ladder wishes.

Will be good when we start using it though... Rather than having over a hundred Word documents being emailed all over creation to wind up in front of one poor guy who has to cut and paste into another document to save as HTML, everyone will just go to a happy URL. Supervisors will have to approve the reports with the click of a button, and everything will be dumped automatically into HTML.

May be the first success for open source in my company, thanks to my "sneakiness..." :)

While I'm thinking of it... My Win2k box at work has been up over 80 days. I'm pushing it for a record and surprisingly enough, it's not letting me down! Of course why shouldn't it be able to handle browser windows, Outlook, PuTTY windows and playing some mp3's? That's about all the box does...

What's nutty to me looking at what I've just typed here, and considering what I've been up to recently, it's hard to believe it hasn't even been a full year yet since my first install of Linux...

As far as personal life goes, the recent dog-related bad/sad news was perhaps just a harbinger, as early last week my marriage came crashing down around my head. Crazy world.

well all of our friends baby / they're going insane now
Butthole Surfers

16 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!