Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×

Comment Re:Oh...they have access to better imagery... (Score 1) 82

Google doesn't (Or didn't until a few days ago) own satellites. They buy their satellite imagery from Geoeye and DigitalGlobe. I'm not sure if they own any aerial providers or if they just buy aerial imagery from a third party. Aerial imagery is higher resolution, but it's a lot easier to shoot down an airplane if you don't want it taking pictures of something.

Comment This Isn't Hard (Score 1) 59

The main thing you need is a consistent source of ephemeris data. I've gone poking around a little with Google and haven't really found a good one. There are two or three NASA services that claim to provide the data, but they're never responding whenever I check. I suppose if you're a bit more motivated than I am (my motivation is currently at a "meh" level,) you could probably find one.

Comment Re:C strikes out again (Score 1) 127

I should point out that the same programmers are usually the ones with tons of null pointer exceptions in their java logs that no one ever reads. You'd think someone would read the log every so often, say "Hmm. There are a lot of exceptions in here. Maybe I should fix a few of them since the stack trace points you right to the line where the exception occurred!" Nope! Not so much!

Comment Re:C strikes out again (Score 1) 127

The programmers always know how big those arrays are. They're just lazy or bad in a variety of ways. It's easy enough to bound a read or copy to a specific size. They just never actually do. I've been on a couple of big C projects and a few smaller ones and the programmer always know what they're working with in a specific function. The problem is never that they don't know how big that specific thing is. The problem is they make no effort to validate the size, or that the pointer's not null, or that someone put good data in it. The number of ways programmers will find to be bad or lazy are countless, and I don't think you can make a language that will account for all of them while still being useful.

Comment Re:Sounds awesome except.... (Score 4, Interesting) 191

If you have a claim to their assets, you could probably still go after all the patents that company owns and presumably any of the licensing fees they had managed to collect. Denying the people pulling those strings as much money as possible still has some value. Now that the precedent has been set, more companies might elect to pursue this course of action, which could put a serious dent in the patent troll business plan. And if you suspect a corporation has just been formed to serve as an alias, you can still go after the assets of the individuals holding that corporation.

Comment Re:I did a contract there briefly (Score 1) 166

Yeah SCO. IIRC they even broke nroff/troff out to a billable component and wanted something for that. I think I sat down and worked out that it'd cost over $10,000 to build fully functional UNIX system. I was never involved in ordering a Solaris system, so I didn't know how the billing for other UNIX systems worked. Most of their clients seemed to wise up within a couple of years.

I forgot about HP and HP/UX, which someone else mentioned here. I guess they're still around, seemingly despite their best efforts to drive their company into the ground in a flaming wreck. The only place I ever saw HP/UX was while working at IBM building printer drivers. It was almost as annoying as SCO to work with.

Comment Re:I did a contract there briefly (Score 5, Interesting) 166

Intel took all the workstation vendors by surprise, but it was their own fault. The prevailing attitude at IBM while they were doing OS/2 was that the PC was a toy and if you wanted to do REAL multitasking, you bought an AIX workstation at a minimum. They were convinced that Windows wasn't going to go far and were positioning OS/2 as a glorified terminal to their larger machines. And it was actually pretty damn good at that, but I digress.

So there we are in 93 or 94, the 386 just taking off, OS/2 and Windows are still pretty much children's toys compared to UNIX and mainframe OSes, the only commercial Intel UNIX is $1200 for the base OS and the fuckers want another $1200 for a C compiler, you can take your chances with a bunch of BSD tapes and I'd just heard about this nifty new Linux thing coming on the scene.

Almost overnight PCs weren't toys anymore and most of the UNIX workstation vendors are going down in flames. In the late '90's I attend a Linux con in Denver. SGI's there, and their marketroid is telling us their company's betting on Windows NT and storage solutions. I didn't have the heart to ask him why I should buy a storage solution from him when I could get one from IBM and know they'd still be there in 5 years. A few months later, SGI declared bankruptcy. Now my phone's more powerful than their old machines.

Of all the old UNIX workstation vendors, I think IBM is the only one left. SGI's still around, of course, they have an office within walking distance of my house. Dunno what they do these days. At least those fuckers who wanted $1200 for a C compiler also went out of business. Damn I hated working with their UNIX. You couldn't wipe your ass without them wanting to charge you for it. That very first slakware distribution that I downloaded onto 26 floppies was better than anything they'd ever done.

Comment Re:Umm .... (Score 1) 209

True, but all our timekeeping math works great if you're thinking in angles and degrees, and a lot of it came about long before geometry was invented or anyone realized that the earth was round and went around the sun. Time makes a lot more sense if you realize that and think Pi instead of nice round numbers.

Slashdot Top Deals

"It says he made us all to be just like him. So if we're dumb, then god is dumb, and maybe even a little ugly on the side." -- Frank Zappa

Working...