There's three ways to build RPMs: the silly way, the easy way, andthe better way.

The silly way starts with /bin/su root , which should be enough of a warning sign. You really don't wantto be building and installing packages as root unless you don't have achoice, and you do.

The easy way is:

$ cat >$HOME/.rpmmacros
%_topdir /some/where
^D
$ cd /some/where
$ mkdir BUILD

One of the things they tell you in programming school is 'side effectsare bad; you should avoid them if possible'. Apparently every now andthen I need a reminder of this, because side effects were at the bottomof my feed whoops from last night.

One of the most expensive bits of DWiki is converting pages from DWikiText into HTML. Originally the conversion was simple andstraightforward; stick wikitext in, get HTML out. But as DWiki grew morefeatures, the wikitext started having additional information: whetherpeople

In the process of enabling caching in DWiki , I managed to illustrate theimportance of thorough testing by failing to properly cache the realtitles of entries and then not noticing, because that the Atom feedsstayed the same was the one thing I didn't test. (I carefully andsomewhat obsessively tested that the rendering of all of the otherpages did stay the same. I even have that automated.)

This caused the syndication feeds for WanderingThoughts to revert toplain filename based titles for eight hours or so, until

This isn't my work; it's from a comment on an earlier entry . I'm repeating it here partly to haveit handy, and partly because DWiki mangled the nice clear plainASCII formatting of the original. So thank you, mostly anonymouscommenter.

The following are the explanation of patchdd script exit codes:

David Heinemeier Hansson in the comments on one of his entries :

Joe: Atom is just RSS without the bugs. [...]

What he said.

The more I've learned about syndication formats, the more thankful I'vebeen that I picked Atom way back when .(I'm not sure why I chose Atom; possibly because it seemed the more upto date of the choices at the time, since it had an RFC in developmentand all.)

The difference between Atom and RSS

When I start grumpily thinking of ways to punish a program's badbehavior is about when it sinks in that I don't like it. Which meansthat Google Desktop has a problem.

Back in December I wrote about Google Desktop and conditional GET and concluded that I didn't have enoughevidence to really know what the heck was going on. Well, I have severalmonths more data and more people using Google Desktop against me, andI've reached a conclusion: Google Desktop doesn't do

Since I alluded to it in passing in an earlier entry , I might as well describe what I know about how THEMIS set up their systems to handle theload. (Disclaimer: this is second and third hand.)

To cope with the visitors to their regular web site, THEMISput their ordinary web servers behind eight Squid proxies , which were in turn behind a load balancerbox. This apparently held up very well to the quite a few millions ofextra visitors from Google Mars .The main movie page was on

Today's random optimization thought, brought to my mind by the processof trying to make DWiki run slightly faster:

If you can't make it faster, maybe you can call it less.

Caches are the degenerate brute force case of this, but programrestructuring is better. Repeatedly calling the same core routine canbe a sign that the upstream interfaces aren't really doing what higherlevels want, so the higher levels have to do redundant work.

Besides, if you can figure out a way to call

The basic volume numbers are that this week, we:

• got 13,909 messages from 248 different IP addresses.
• handled 18,580 sessions from 927 different IP addresses.
• received 195,437 connections from at least 36,864 different IP addresses.
• hit a highwater of 16 connections being checked at once.

These are mostly down from last week ,although not as much as I would like to see. Again, the per