According to Jeremy, Google is making some sort of concerted effort to de-value blogs in the determination of what matters on the web. This raises an interesting question – what matters on the web? Since the web was created ab initio to enable us to link from information to more useful information, sure, the stuff that we link to is, in fact, what matters? The fact that Google results change every day is a good thing, not a flaw. What matters changes every day also. That’s why folks watch the news every day, not just once. It will indeed be a great disappointment if he is correct in his assessment. While I don’t particularly care what algorithm they use, the concept itself – stuff that lots of people link to is stuff that matters – makes a great deal of sense.

The perceived danger is very understandable – momentary blips in interest in a particular thingy may cause search results to get skewed that direction for short periods of time, and the semi-permanence of blog archives may make that skew permanent. So then one has to figure out if there is a “correct” answer to a particular search term. That is, is Google an encyclopedia index, or a web search tool? With my loathing of folks letting Google be their research assistant (must elaborate on *that* some day), it seems to me that it should stick to what it’s good at — being a web search tool. And since linking is what the web is all about, if Google drops this aspect, they will quickly lose their top dog status to someone else who will give us what we want.

Thanks mostly to a large gift that *someone* gave me early on, I have just cleared my first million in the blogshares game. And, in the spirit of his generosity, I’ve been spreading the wealth a little bit also. It is indeed a lot more fun to play the game when you have a ton of cash to sling around. But, then, I suppose that’s true in the real world also. 😉

So, if I understand Chris correctly, the main complaint with Trackback, compared to Pingback, is a complaint with a particular implementation of it, not necessarily with the concept itself. As far as I can tell, MT does the two things that Chris says that Trackback does not do. Viz, ping URLs which are just mentioned (linked to) in the body of the article, and be able to ping more than one URL with a single article. At least, that’s certainly the way that I understood it when I was setting configuration variables.

What I find irritating is that there are two incompatible ways to do this VERY simple thing, ensuring that no matter which one I choose, I can never be sure that my pings will be acknowledged.

Then again, that’s the nice thing about standards …

So it turns out that, as people have been telling me, the MT software license is in fact pretty not-open. Among other things, it says

You may not redistribute the Software without Licensor’s prior
written consent. Although you may modify or create derivative copies
of the Software for your own use, you may not distribute modified or
derivative copies of the Software.

I suppose that’s pretty normal for software, but does not mesh with the way that I view the world, and, I suppose, it’s pretty clear that I’ve already violated it, having distributed modified versions of various parts of the software. So I suppose I’ll have to find some other package to migrate to. I’m leaning towards Ken Coar’s software, if he lets me use it. I expect he’ll have a somewhat more open license,

Evidentally, some people don’t have trackbacks set up correctly.

I spent all day yesterday, and some of the day before, trying to migrate services off of a dying server before it completely died. Mail, DNS, CVS, and a plethora of test and demo vhosts inhabited the server, as well as my news and blogs site.

I was getting periodic kernel panics, that were coming with greater frequency as I tried to move services. In typical saved-at-the-last-minute movie fashion, as soon as the last bit of data was recovered, the server crashed, and will not even so much as power back on. Although I’ve managed to mount the hard drive in the new server in case I need to recover any more files that turn up missing. The new server is one of the machines that I bought for my training room, but, with training going at the rate that it is right now, I don’t think that it will be missed. Added a little ram, and now I have a $500 server, replacing one that probably cost $2000 new, and has had many times that much spent on it in maintenance time. (Hi, sungo!)

So, chimp is dead, long live barolo. I hope it can run for 2 or 3 years without much needing to be done to it.

Our LUG has put together a simple library thingy for our web site, which lets us list the books that we have in our library, and lets people post reviews of these books. This, in turn, encourages the publishers to keep sending us more books. (Lest anyone should think, for a moment, that this wasn’t entirely mercenary in motivation.)

The software is mod_perl, and, now that it is mostly functional, we’d be delighted to let people poke around at the source. We had talked about moving it to SF.net if there seemed to be any actual interest, since, at the moment, the cvs repository is on my home machine, which makes me somewhat reluctant to hand out accounts like candy.

Yes, it is very limited, annoying in a number of ways, and lacks some rather important features. But we are making good progress, and hope to have something a little more functional pretty soon.

I’m somewhat baffled by the BlogShares phenomenon. It’s like a virtual stock market, where one trades blogs. But I’m really unclear how the valuation works. I just managed to sell 10 shares of my stock for a whopping $0.00 each because the interface is so amazingly unintuitive.

Evidentally RoUS’s blog is a member, as is sungo’s. And, yes, that was just a shameless opportunistic linking in the hopes of driving up my valuation, but I don’t really pretend to know if that will work.

*sigh*. I don’t understand the real stock market either.

Having gone an inordinate amount of time without doing any real programming, last night krietz and I hacked on some mod_perl stuff until about 11. It’s a little package to allow our LUG to keep track of what is in our library, let people check out books, and write reviews of the book.

So we got stuff working at a rudimentary level, and now I’ve done some tweaking today. It’s pretty slick, and soon we’ll be ready to unleash it on the world. May even make it available for use outside of our little world if there seems to be any interest.

The only really disappointing thing about this project is all the people that said that they wanted to participate, and then never did. I know, there are always good reasons for this, mostly having to do with people being so darned busy, but I was really looking forward to doing a collaborative project with some of these folks.

Perhaps I’ll actually get some stuff done on the Malibut project, and get to participate in a collaborative project from the start. That could be a lot of fun.

I have, for a long time, mocked Linux for the absurd difficulty of setting up printing. Well, I have spent much of the last 3 hours trying to get Windows to print to a perfectly-working printer hanging off of my Linux machine, which every other machine on my network can print to. I’m not sure if this is Samba’s fault or Window’s fault, but, really, how hard could it be for Windows to have implemented Unix printing decades ago so that I would not have to be suffering this tormet right now. I’m not amused.