The worst and the best of cycle racing

I finally got around to watching the TiVo’ed coverage of last weekend’s “Amstel Gold”, one of the “Spring Classics” pro bike races. Unusually for them, the weather was beatiful – probably too hot for the tastes of the riders, but it made for good coverage and I’m sure the fans appreciated it. In previous years it’s been wet or snowy or so foggy that the tv coverage was almost non-existant.

As usual on these sorts of races, there wasn’t much happening until the last 30 minutes or so. There was a break-away group up the road and a big peleton, but a couple of the race favourites manged to bridge up to the leaders. That’s what I mean about “the worst”. 2 hours of watching a bunch of guys cycling without any changes in leadership, without teams organizing chases, without anything really interesting happening. Yawn.

With only a few minutes to go, the lead group consisted mostly of guys who had a reasonable expectation or hope of winning, because of previous wins on this or similar races. The only wild factor was that one of these favourites, Davide Rebellin, also had a team mate with him. I expected this would mean that his team mate, Stefan Schumacher, would attempt to launch him on a break-away on the second last or last climb of the day. But instead, their team played a very clever card.

Schumacher attacked alone. The other guys in the bunch wouldn’t counter attack to bring him back because none of them wanted to tire himself out and give the upper hand to one of his rivals. You could see Paulo Bettinni and Michael Boogerd trying to get the others to lead the counter attack. They just couldn’t get it together to cooperate, knowing that Rebellin would sit on any counter attack but wouldn’t contribute to it. So Schumacher sailed on ahead and won by a good margin. Even better, Rebillin used his tactical advantage to grab enough rest that he could outsprint the rest of the group to take second. And that’s the best of bike racing, the team tactics that say it’s better for some second banana in your team to get a clear win than for your team leader to fight it out in a bunch sprint. Every team has its star, but when it comes right down to it, it’s the team that matters.

Ok, that’s one thing

I discovered why OpenID comments went from working some of the time to working never: last time I upgraded my blog I’d inadvertently made it so none of the directories within my blog site were writable by the web server, and the OpenId plugin makes temporary files. That lead to the discovery that as well as OpenID comments, the directory used by the Gravatar cache wasn’t writable, which lead to the discovery that the Gravatars weren’t updating at all, even after I fixed the permissions, which lead to the discovery that when the Gravatar plugin discovers it can’t write to the Gravatar cache directory it just silently turns the “Enable caching” option to off, which lead to the discovery that even when I turned it back on, it still wasn’t updating, which lead to the discovery that there is an option in the plugin to “Use REST protocol” which evidently Gravatar doesn’t support any more so I had to turn that off. Phew. Now Gravatars seem to be working again.

Knees still hurt like hell, though.

META: Don’t use OpenID

OpenId comment authentication seems to be extremely hit-or-miss. It works for some, and for others their posts get flagged as spam, and for others still they get swallowed entirely. I’ve tried to debug it, but I haven’t figured out what’s wrong. I tried to deactivate it, but it just made things worse. So until further notice, please don’t use OpenID or your LiveJournal ID to comment on here.

Update: I just remembered that I had to hack the source in order to make this work before, and I recently installed an update. The update probably over-wrote my hack. Now to dig though the backup to see if I can find the hacked file.

Oh, this bodes well for meeting our end of month deadline

The ClearCase server machine is still dead. Evidently both hard drives went tits up yesterday, and nobody can bring it up. And I’ve got a bunch of files not checked in – from what I remember back when we had a ClearCase administrator, files that aren’t checked in aren’t backed up.

And just to make my day complete, I can’t log into Lotus Notes.

Good thing I’ve got some movies on my iPod.

ClearDDTS? I have to ask…

Was there ever in the history of the world a worse bug tracking system than ClearDDTS? The user interface is so ugly, inconsistent and unfriendly that the only way they could have made it worse is if they’d used Lotus Notes as the front end. Slow? It takes a good 30 seconds to enter a new bug or process a change in an existing bug, something that should take way, way less than a second in any competent relational database.

And to top it off, recently it’s taken to kicking everybody off for 5-10 minutes at a time and not letting you log back in, while the sysadmin says “there isn’t anything unusual in the log files”.

The only reason for not throwing the whole thing away and switching to Bugzilla or something like that is that it integrates so well with ClearCase. And it has become apparent to me over the last few weeks that the only reason I loved ClearCase is that we had a really good full time ClearCase administrator, Steve. Since the powers that be fired him, we’ve got two guys struggling for a week at a time to do stuff that Steve could have done in a few minutes. And I no longer love ClearCase.

Let’s throw away the whole thing and replace it with that CMS that integrates with Subversion, ok? What’s it called, Trak or something like that?

Update: Just got an email from the sysadmin – evidently a process keeps dying. One called “update_htpasswd.sh”, which evidently updates the .htaccess file based on what is returned by “ypcat passwd”. Ok, I’m not an expert, but aren’t there ways to authenticate directly to NIS rather than building a .htpasswd file at regular intervals? Sheesh, talk about amateur hour!

Update 2: Minutes after posting that, the ClearCase vob crashed as well, and so I left for home.