What's it all about, Alfie?

I've been working exceptionally hard these last few days, and it should come as no surprise that it's a thankless, horrible job, but they saddled me with it and rather than drag it out, I want to get get this crap done, and move on. Unfortunately, this will probably mean that I'll be rewarded with even more horrible and distasteful work on this project in the future. I suppose on one had I should be happy that in these times I have a well-paying job in the financial sector. But I tell you this... I'd easily work for less to do something that wasn't as distasteful.

Even this place, a year ago, was a better place. But this Death March project that I'm being forced to be on is just draining my will to live. It's bad. But I'm trying my best to stay optimistic - tonight I'll get more than 4.5 hrs. sleep, and that will be a welcome relief.

I've got people looking, but so far nothing really exciting where the work is at least as interesting as this, and is a long-term commitment. No more short-term consulting jobs. I just sure hope this horrible crap is over soon.

This morning I saw that Perian 1.1.1 is out, and I tried to get it. First, through the auto-updater in Perian itself, and then directly from the Perian web site. I was stunned... it took me several hours to get the update, and in the end the auto-updater was the 'winner'.

I was stunned. I've never seen a web site under such load. I may be trying to infer causation in a correlational study, but if they are getting hit this badly from an update, they might need to get a better hoster - or maybe get more donations because this thing is big.

I'll admit, I probably never see it in action - QuickTime just works and the plugin is just that - a plugin. But I have to say I'm impressed. Go get 'em guys!

There was a minor bug in the saved passwords code that caused some users not to be able to retrieve them, and that caused the developers to release 3.0.3. If I were impacted, I'd have wanted a fix, but as it was, I wasn't and things were fine.

As an aside, I've gone back to using the straight Firefox as the guy building optimized versions for the different processors is at 3.0.1 and if he catches up I may try to compare the two, but I have a feeling that it's like the MacVim tests I've done - on my machine the difference might very well be next to zero.

This morning I got word that the MacVim maintainers released Snapshot 35 with several bug fixes and added features including faster drawing. I had to get it right away.

After some testing I'm still a bit surprised that the ATSUI renderer is not proving to be any faster on my MacBook Pro than the default renderer. My last theory that they are both being backed by the same Core Graphics methods would lend weight to the non-difference, and that might be effected by the graphics card being used. So maybe on a slower box or worse graphics card, the difference is clear, but on mine they are dead even.

Not bad... just interesting. I love Vim.

One of the support crew in London pointed out to me that a Money Market ticker in one of the applications fed from my ticker plant was not updating properly. That the quote from my source had both the bid and ask less than zero was interesting. What happened was the conversion I did from working in "all positive" price to "mixed sign" prices on the calculation of 'Best Price' in the client code had a bug. Basically, there is a test for a 'good' quote that the bid and ask are in the right relation to one another - with a little 'fudge' factor, and also that they are not horribly disparate either.

The problem code looked like:

    if (((fabs(mAsk) <= 0.5) ||
         (1.1*fabs(mAsk) >= fabs(mBid))) &&
        (fabs(mBid)*1000.0 > fabs(mAsk))) {
      retval = (mBid + mAsk)/2.0;
    }

where the firs check is to make sure that if the ask is small enough we don't mess with the check - like 0.05/0.10 quotes. The second check is the problem... without the fabs() calls, it works for positive numbers: of the ask is bigger than the bid, and the ask isn't too much bigger than the bid, then we take the mid point. Easy.

But if we put the fabs() calls in there and try to work with negative numbers, we're messed up. Negative bids need to be numerically less than the asks but with the fabs() calls, they get flipped to being numerically greater than the asks. This is the problem.

If we change the code to look like:

    if (((fabs(mAsk) <= 0.5) ||
         ((mBid - mAsk)/fabs(mAsk) <= 0.1)) &&
        (fabs(mBid)*1000.0 > fabs(mAsk))) {
      retval = (mBid + mAsk)/2.0;
    }

then we're looking at the percentage change and that works for both signs - positive and negative.

I put this into the clients and things started ticking nicely - if you can call a negative price quote 'nice'. It was just interesting to me that I thought I had it right the first time, but you have to be very tricky about throwing around fabs() in your code.

This morning I read on the Wall Street Journal front page about the failure of WaMu being the biggest in the nation's history. Roughly $30 billion in debt and stock washed down the tubes. Holy Cow!

The fact that no bank was willing to buy WaMu until it failed shows how badly confidence has eroded in a banking system awash with record profits just a few years ago. Faced with deepening losses on mortgages, credit cards and other loans, big and small banks across the country are struggling with what many bank executives say is a crisis far deeper than the savings-and-loan debacle.

This just makes me mad. People will look at this industry and say "What a horrible bunch of greedy people" when the last statement in the quote says it all: bad loans.

OK, who is the major borrower here? Answer: People. People getting into homes and mortgages they can't afford. People running up credit card debit they can't pay. It's people. Every single one of us that lives beyond their means. Ask not for whom the bailout is - is it for thee

I got an email from the great guys at Panic that they have created a new Unison (NNTP) server farm to replace the old one. The old one was their own after they stopped using a service that did a good deal of filtering - all to the good. This new server farm is said to be bigger, faster, with more groups and longer retention.

One of the nice things they added was SSL access - built right into Unison. I like that only because sending plaintext anything these days is asking to get snooped or swiped. It just makes the connection a little more 'private'.

Love the work these guys do.

I was looking at iTerm again this morning as a new version was released and I had forgotten that they had an option to remove the vertical scroll-bars on the windows and still have a scroll-back buffer activated by the multi-touch gesture on my MacBook Pro. This was a nice feature. So I thought - I wonder if it's possible with Terminal.app? Answer: No, not really.

I googled a lot of hits and all said they got it working, but I think they got it working on a previous version of Leopard - not 10.5.5, because everything they mentioned that 'worked' I tried and it didn't work. Not a bit. I looked into the nib file and no lock there, but I think I got a hint at the reason it's not possible any longer: Terminal.app allows the user to resize the window.

If you have that working, then there needs to be some 'area' or buffer on the right side of the window where the resize gadget fits, and if you try to remove the scroll bar, you run into trouble. Firefox 3.0.1 had this problem on Liza's MacBook Air - you need the status bar at the bottom to 'hold' the resize widget or you get into trouble. So the solution would have to be that if you didn't want the scrollbar, you'd have to remove the resizing widget and then the user may not easily know how to resize the window.

But that seems reasonable - and from what I'm hearing on the Google hits, it was working. It's too bad they took it out. Oh well... maybe in Snow Leopard they'll put it back - or at least use a consistent set of scrollbars for all windows.

I'm tired of companies selling a service, getting the money, and then coming up with a million diffeent ways to keep for providing that service. Comcast is just like that. Insurance companies are like that as well.

Comcast is in the news again with a statement about how the 250GB/mo. limits will raise as general network usage rises in the future. While that's true, so will the monthly charges - it's not like they are planning on giving us more for the dollar - even in the future. It's all another shell game.

Comcast needs to realize that most people don't need all the bandwidth they have all the time and be the Corporate Good Guy and make the system capable of handling the load. Be the one people want to have. Then you've got public opinion on your side if the regulators get ugly. They're already making a ton off the TV market and you can only watch one show per set - two if you think you can multitask that well.

Stop trying to find ways not to provide the service.

Different folks, I know, but a nice pair of updates from the Mozilla Camp - Firefox 3.0.2 and Camino 1.6.4. The Firefox update seems to be focused on the security angle - good enough, and the browser wasn't bad to begin with, but they seemed to have squeezed in a few nice things. Several additional languages have been added, some fixes for the right-to-left languages, and even some Mac-specific bugs fixed.

Gotta get it, folks. Great browser.

Camino 1.6.4 is really the surprise for me today... it's got the latest rendering engine and boy is it improved. The previous version wasn't even worth using for more than a few minutes. This is fast enough now on pages to actually use. I'll be playing with it in the weeks to come, but if they keep making progress like this, I may move from Firefox as my #2 to Camino. It's really kicked in the speed. Great work, guys!

After reading up a little more on Camino, I have to say it seems like I'd like these guys. They are serious about making nice Mac software on the Mozilla codebase. Can't beat that. They have tried to incorporate the Safari-like Find interface, and several other nice things I use to make it a natural conversion from Safari to Camino. Yes... if the speed holds up, I may switch to Camino as my #2. Nice work.

UPDATE: I'm sad to see that the buttons in Camino are a mess. Maybe it's just the few web sites I tried, but the text was horribly off-center and the buttons barely looked like buttons at all. Not good, sad to say. But getting better, little by little.