I use Interbase and am always amazed at how little attention it gets compared to MySQL and PostgreSQL. It has a lot of the more advanced features people claim aren't available in open-source databases, e.g. the online backup capability the article went on about. It had transaction support from the get-go, supports stored procedures, and on and on.
It baffles me that the author of the article could know about Interbase and yet seemingly not even look at it in the course of researching the article.
The articles in this particular story make pretty poor examples of what you're saying, though. The AnandTech review in particular isn't flattering at all -- as far as I'm concerned he made the new card look awful compared to the GeForce3. Seemed like a fair review to me. Tom's Hardware was somewhat more forgiving but still far from a puff piece. Both of them came straight out and said, "This card isn't ready for prime time yet."
Of course, we'll see if they get to review ATI's next new card, but it's clear that the so-called catch-22 isn't universal.
Then why hasn't it happened yet? Surely there were valuable patents issued before 1981. If it's so trivial to get patent terms extended, why didn't the holders of all those pre-1981 patents do it?
Copyright law and patent law are two different beasts, though they both cover intellectual property. In the case of patent law -- which, keep in mind, also covers physical inventions -- there are economically significant players who stand to lose a lot if patents are extended indefinitely. It's not like copyright law, where the main opponents to extension are librarians, academics, and well-meaning citizens. There would arguably be well-funded corporate opposition to extending patent terms. (I agree there'd also be well-funded support for the idea, but it's not as grossly one-sided as is the case with copyright law.)
US patent terms have changed a lot less frequently than copyright terms, historically speaking. They changed from 14 to 17 years in 1861. They stayed that length for over a hundred years, until a 1994 amendment passed to implement the Uruguay Round of the GATT treaty, which changed the term to 20 years from application date.
There have been plenty of big companies with valuable patents over the years. To my knowledge there has never been a concerted lobbying effort to get patent terms extended; if there has been, it hasn't worked. One reason is that it's much easier for a legislator who isn't versed in IP law to see the social benefit of patent expiration than of copyright expiration. From lug nuts to tumbler locks, you can point to any number of real-world examples and easily get a lawmaker to agree that it would have made no sense for one company to own those technologies till the end of time. It's a much more abstract, fuzzy argument to say that it's good for society if Disney stops owning Mickey Mouse. (Which, I hasten to add, I think it is.)
If the anti-software-patent side loses, yes, it will be annoying as hell, but the one thing that seems to be lost in a lot of the rhetoric (on both sides) is that patents expire. The people filing bad patents will get a big win for 20 years; then nobody will ever have to worry about violating their patents again. It'll get harder and harder over time to come up with simple, never-patented-before ideas that inconvenience large numbers of programmers. Patents will tend to get increasingly specialized and obscure as time goes on. Things are messy right now only because the whole idea of patenting software is so new that many of the fundamental building blocks are vulnerable to patents. As the building-block patents expire, people will immediately swoop in to take advantage of that.
In fact, we're already seeing it in a few cases. RSA's public-key patents are an obvious example. Three years ago a lot of people were livid about the restrictions those patents placed on developing crypto software. Now the patents are over and we have software like OpenSSL which are better than the commercial alternatives. (Granted, development of much of this software started long before the RSA patents expired, but the point is, nobody is worried about these supposedly horrible, disastrous patents any more.)
Obviously 20 years is a long time in the software biz, but the point is, it's not forever, and the situation will improve over time even if the bad guys win.
Some of these have been mentioned already, a couple not, or maybe they've been modded way down.
Dark Angel. No, not because Jessica Alba is a hottie. It has wonderful continuity, good interplay between the major characters, a truly despicable villain who nonetheless isn't just a one-dimensional cliche, and a visual style that isn't like anything else on TV right now (even if it's maybe not entirely original). It also makes a decent attempt to be plausible in the science department, most of the time anyway.
Futurama. Continues to do a hilarious job of skewering sci-fi cliches.
Andromeda. Again, good continuity with a sense of a larger story being told (as a B5 nut this scores lots of points with me), interesting characters, good effects, and story concepts we haven't seen done a million times before.
Lexx. This brings me back to my college days of watching videotapes of awful old sci-fi movies on Saturday nights. Last season kind of dragged on but this year is a hoot so far.
Stargate is just below my threshold -- not bad and I catch it pretty often, but I feel like I watch enough TV as it is so I don't follow it religiously.
Depends on how you define "phone," I guess. If you're talking the typical tiny-as-can-possibly-be Motorola, then I agree, it's kind of hard to see the point. But if instead you're talking something more like the Kyocera SmartPhone, I think it's a little less crazy.
To my thinking, the power and flexibility I'd want in a mobile device is pretty much proportional to the richness of interaction I can have with it. Cracking the rich-interaction problem on small mobile devices, it seems to me, is going to be a much tougher job than cramming in software and CPU power that used to only be available on the desktop.
I do a few things that are extremely effective in filtering out spam. I have procmail rules to do the following:
Mail that doesn't list one of my addresses, or the address of a mailing list I know I'm on, in the To: or Cc: lines gets filtered. This alone catches a solid 85-90% of my spam flow, though it seems to be getting less effective as time goes on.
Mail that's from a free E-mail service (Hotmail, Angelfire, etc.) gets filtered.
Mail that contains certain keyphrases (e.g. "free" in all caps, or "this is not spam" or "S.1618") gets filtered.
Mail that has passed through a.cn or.tw or.kr host gets filtered. Those countries seem to have an abundance of open relays. At some point I hope to change this to check against ORBS/DUL instead.
Now, the interesting thing is what I do once I've decided to filter the mail. Since my rules catch legitimate mail, I don't just throw it away. I wrote a small collection of Perl scripts (which I'll release to the world someday soon, but they need documentation) that maintain a whitelist of sender addresses.
If a filtered message is from an address that's marked valid, it's delivered. If it's from an address that's marked invalid, it's discarded. If it's from an unknown address, the message is put in a holding area and an autoreply is sent back to the sender from a magic address asking them to reply in order to validate themselves.
The magic address is unique per filtered message -- it uses qmail's address extension mechanism -- and mail to the magic address never gets delivered to me, so I don't care if it gets added to spam lists. The Perl script behind the magic address does a quick check to make sure it's not processing a bounce, then marks the sender of the original message as valid and delivers the original message (or messages if more than one arrived while awaiting validation).
Held messages are cleaned out by a cron job when they get too old.
This is sort of similar in concept to the password mechanism of SpamBouncer or (a closer cousin) SpamCop's whitelist feature, but it doesn't require senders to retransmit their messages, which I always thought was pretty annoying to ask people to do since not everyone saves their outgoing mail. Granted, asking them to do anything is kind of annoying, but at least this is less so since they can just hit "reply" and "send".
This setup is cool because it allows friends to Bcc me on stuff without my "I must be listed as a recipient" rule trashing their messages, even if they've just switched E-mail addresses. It is admittedly based on the assumption that spammers don't read replies to their mail and/or wouldn't go to the effort of unlocking themselves; I have yet to see a spammer do that, and given the economics of spamming I think that'll be a safe assumption for the foreseeable future, unless this approach gets so popular that spammers start writing automated unlock bots!
I agree with all of that. But it's also possible to ask good interview questions, and I think I do for the most part.
Let me give you one of my questions from the story in question, which alone eliminated a good 40% of the candidates.
"In a SELECT statement, what's the difference between the WHERE clause and the HAVING clause?"
For those reading this who aren't SQL experts, that's vaguely equivalent to asking a Perl programmer what the difference between an array and a hash is. Yeah, it's possible to use Perl for years without knowing that, but it's a fundamental enough part of the language that if you can't answer that question you shouldn't claim to be an expert, and further, if you can't answer it and you do claim to be an expert, I become very nervous about how accurately and truthfully you've gauged the rest of the abilities you list on your resume.
I've asked interview questions I don't know the answer to, but never without clearly stating that fact in advance, and those questions were always high-level algorithm design, "see how they approach the problem" types of things. For example, at one point several years ago I was playing in my spare time with writing a program to lay out crossword puzzles (not as simple a problem to solve efficiently as it might intuitively seem) and I asked lots of candidates how they'd approach that. I was mostly looking to see what ideas they came up with and discarded as they thought the problem over. It was not important which specific approaches they tried; I wanted to see how they arrived at those approaches.
When I need to ask specific design questions, I always ask the candidate to solve some problem that I or one of my colleagues have just recently solved ourselves, again so I can see how their thought processes compare to our own. Which, BTW, also provides insight into how well a person will mesh with the rest of the team, though it's not the major way I judge that.
I've never worked for a company where I was expected to let HR grade someone's technical skills. That sounds like a recipe for disaster. You can't ask anything beyond extremely rudimentary questions and expect an HR person to make sense of the answers, especially since, as you say, looking at the person's thought processes is critical.
If he'd been an employee, that would more likely have been my approach. In fact I've always fought to let my people work on new kinds of projects, spend time fiddling with side projects, etc. to expand their skill sets. But the bozo in question was a contractor, and it hardly seems reasonable to pay him by the hour while he learns a language he claimed to have already known when we interviewed him.
I was a manager during the period in question, and I can tell you, I had a devil of a time filling a couple of positions. Let me take issue with this statement from the article:
If employers were that desperate, they would certainly
not be hiring just a minuscule fraction of their job applicants.
I'll take one example. We had a data warehouse (mostly a big Oracle PL/SQL application). The engineer who designed and implemented the original code left the company, and I was tasked with hiring his replacement since there were some pretty substantial architectural changes we needed to make.
Now, there are a lot of database people out there. A lot. I looked at more resumes than I can count. My ear was sore from phone interviews. Thing was, just about everyone I talked to fell into one of two categories:
Listed all sorts of Oracle skills on their resume, but couldn't correctly answer my SQL skill-testing questions. (Which were nothing especially complex.) This was the vast majority, which surprised me.
Knew their SQL, but clearly had no design skills to speak of. I'd ask them to design a trivial application and they'd either botch it or claim that they just wrote code, someone else always designed it and gave them a spec.
I looked and looked. The executive staff got really antsy and started leaning on me to do what the article suggests, just hire someone to get the work going, even if they weren't perfect for the job. I resisted for a while but finally caved in.
The contractor we brought in -- one of the better ones I'd interviewed, though I hadn't liked him well enough to want to hire him -- did a decent job of talking to the right people, gathering requirements, and getting himself acquainted with the layout of the code. But then he started to submit his own code, and man, what a disaster. I wasted weeks correcting his mistakes. Finally I fired him and went back to my original search.
The specifics of the story here aren't important. The point is that it doesn't take many times being burned by the "hire any bum off the street, just fill this technical position" attitude before you develop a very healthy caution about hiring the wrong person. I've seen it happen at other companies and I think it's a universal truth: hiring the wrong person for a job can leave you in a much worse position than hiring nobody at all. Not least because you think you have the position filled, so you stop looking for a while.
Experienced managers know this, so they put themselves through the "there's nobody out there!" routine when the job market is tight. It sucks massively, but it sucks less than the alternative.
(How did the story end? We found an H1-B person who fit the bill perfectly. Then the government took so long to process his paperwork -- months -- that by the time it came through, he'd gotten cold feet. Ugh! Happily by that time I'd moved to a different group.)
Couldn't agree more. In fact, I liked it so much, I helped write it!:) (I wrote the video 2-frame deinterlace algorithm and some of the automatic aspect ratio detection code, among other things.)
For those who don't know about it, I'll expand a little bit: DScaler and a TV capture card takes input from a regular NTSC video source (cable TV, VCR, TiVo, DirecTV receiver, whatever) and applies a bunch of video processing algorithms to remove the visible artifacts caused by the fact that NTSC is an interlaced video standard. It uses your PC's video card to scale the image up to whatever resolution you like. It'll even do fancy tricks like displaying film material at an even frame rate if your refresh rate is a multiple of 24Hz (the frame rate used for most film material), or automatically detecting letterboxed movies and expanding them to fill a widescreen display. Lots of people are using it in place of expensive standalone video scalers to display analog TV on their HDTV sets. If you have a clean source signal you can get a picture that looks nearly as good as standard-definition digital TV. Not HD by any stretch, but it still looks quite good.
Best of all, it's GPLed.
The main place where it's discussed, and where all the developers hang out (and a great place to discuss using PCs for watching TV and movies) is AVS Forum's Home Theater Computers message board.
The protests are at least getting a little bit of mainstream press coverage, which I hope will at least make the broader public aware that there's an issue here. Without the support of a bigger slice of the voting public than Slashdot's readership, there's no way enough pressure will be brought to bear to get the DMCA repealed. (Though of course court challenges are also a possibility.)
It's not difficult to check for over-the-air DTV availability in a particular area. Or to check DirecTV's Web site and discover that they don't carry HDTV other than HBO and a few pay-per-view movies. The guy who wrote the article clearly should have done a bit of homework before blowing 7 grand on a video system. It seems like he saw the word "digital" in a few different places and assumed it was all the same thing.
That said, for those of us in places like the San Francisco Bay Area, which has a large number of digital stations, DTV and HDTV are just lovely. On a clear day, my rooftop antenna picks up six or seven digital stations. The picture quality is stunning even on the standard definition stations, much crisper than the clearest cable channels and most DirecTV channels. And HD shows look better than the picture at the local movie theaters. The picture has yet to fail to elicit a "wow" when I've shown it to people.
And the cool thing is, it's on my computer using an ATSC tuner card which means I can record the digital signal to my hard disk for later viewing - not as slick as a TiVo, but adequate. (And before you ask why anyone would watch HDTV on a 17" monitor, the monitor on that PC is one of these, more or less, less expensive than a new HDTV if you buy it used.)
I do wish the prices would come down on more traditional HDTV sets and that they'd get the integration issues straightened out so a separate settop box wasn't required. Better market penetration will equal more incentive for the networks to produce more HD shows. But if you're willing to actually learn about what you're buying, the technology is out there and working.
asking users to pay in order to contribute to an online resource is basically the dumb idea that Napster's currently having.
On the other hand, in many ways it's also the dumb idea behind Ultima Online, EverQuest, Asheron's Call, and other online games, some of which are making a tidy profit. By and large those games are pretty pointless without the other players. I think this is proof that people are willing to pay for the privilege of interacting with strangers, especially if the interaction is of a sort they can't easily get elsewhere.
The problem with Napster's pay-to-contribute model is that the contributions are, for all practical purposes, anonymous. Nobody will ever say "Oh, there's Joe, his ripped MP3s are the best!" I doubt one Napster user in 1000 paid the least bit of attention to who they were downloading from.
Throw in the power of ego gratification, and getting users to pay to create your content for you isn't quite so impossible.
Granted, there are idiots out there using SpamCop when they shouldn't. Doesn't mean using it for its intended purpose is bad. I use it because it's too time-consuming to go through the headers of zillions of spam messages by hand.
As a mail admin myself, I've gotten my share of bogus SpamCop reports, so I can sympathize with what you're saying, but I think SpamCop's good outweighs its bad.
Re:A few discussion points.
on
ORBS Forks
·
· Score: 1
If time=money, why not simply take the quick route, delete those spams, and get on with your day?
Because that's treating the symptom, and we'd rather get rid of the disease. If we can eliminate or greatly reduce spam with our tinkering and complaining -- and I know my tinkering has gotten rid of the majority of the spam I used to get, though I still get too much -- then we save that much time every day for the rest of our lives.
Because spam is about more than the two seconds it takes to recognize and delete the mail; it's about the irritation we feel every time we have to do it. Not a major irritation in isolation, but any minor irritation becomes a major one if repeated frequently enough.
And because we feel it's wrong to spam, plain and simple, and speaking out against what one feels is wrong is, we feel, a right and a duty in a free society.
I use 100Mbps on my home network. 10Mbps doesn't cut it if you're doing any work with video files, for example, or doing big builds over the net. What actually prompted me to upgrade was how long it was taking me to copy my ripped MP3 files to my fileserver, if you must know.
That article had nothing new to say, and what's more, some of their approaches aren't state-of-the-art in silent computing. Some of the technologies and approaches they didn't touch on:
Liquid cooling. The Koolance and CALM System cases are the obvious examples, but for higher cooling capacity other folks have built liquid-cooling solutions that require few or no fans. (Pump noise is easier to deal with since you can enclose a pump in an airtight space.)
External fans. Essentially, this approach is "don't eliminate fan noise, just put it somewhere else." An air duct or tube runs to the PC, and at the other end, possibly off in some entirely different room, is the fan.
Fanless power supply. There are (generally fairly low-wattage) power supplies that dissipate their heat by convection. Typically you'd use one in a case where it's practical to cut or drill holes above the PS. I've seen talk, though not examples, of people putting two of these into one PC to get around the limited wattage; no idea if that'd be practical.
High-wattage power supply. Get a 550W power supply with a temperature-regulated fan to power your low-wattage PC. The fan will tend to run at low speed since you won't be running anywhere close to capacity.
Voltage reduction on fans. This works equally well for case fans, CPU fans, power supply fans, etc. The typical quick-and-dirty technique, though it's not ideal, is to splice in the 5V lead in place of the ground, effectively reducing fan voltage to 7V (assuming it was a 12V fan). The fan will spin more slowly, generating less noise.
Underclocking. If you don't need a tip-top performance PC, reducing the CPU clock speed cuts down on heat. If you underclock far enough you don't even need separate CPU cooling at all. Same goes for graphics cards and, to a lesser extent, motherboard chipsets.
Just to name a few. And of course you typically want a low-RPM hard disk (though the new quiet Seagate drive sounds promising), low-rotation CD-ROM drive (use a drive with Zen's TrueX multibeam technology and you'll still have fast reads), etc.
Apple's iMac Cube is proof that it's possible to build a computer with no fans whatsoever. Hopefully it's a harbinger of things to come.
I used to think that, but I've changed my mind after using the latest version for a while. If you haven't tried it since before the 5.x release, give it another try. Opera's window management, especially with the gesture support in the latest version, has significantly increased the number of Web pages I see during a browsing session.
Right-drag on a link with a down-and-up motion and it opens the new page *below* the current one in the list of windows so you can keep reading whatever page you're on; when the new page finishes loading its entry in the window list changes color.
This is great for Slashdot; by the time I'm finished browsing the top few comments, I can switch (again, using nothing but mouse gestures) to the story being discussed, no twiddling my thumbs waiting for some Slashdotted page to load, or worse, switching back and forth among three windows checking to see if two of them have finished loading yet.
Your point about multiple monitors is a good one, though. I only have the one monitor so it hasn't been an issue for me.
Opera doesn't deal well with things like drop-down menus implemented using CSS and JavaScript, and it's a bit crash-prone (though less so than Mozilla), but it's my browser of choice by a long shot.
Past articles have mentioned some since removed draconian terms of use and there have also been reports of security breaches in the past.
Both of which, it seems to me, are kind of shaky complaints. If the current terms of use are troubling, that's one thing, but the fact that the bad bits have been removed means that the bad bits aren't there to complain about any more. Yeah, they could add the bad bits again in the future, but no more so than any company could modify their terms of use after the fact (e.g. Microsoft could modify the MSDN site terms of use in exactly the same way).
And while I certainly agree that Passport is a much juicier cracking target than the MSDN user database, the fact is that any personal data you supply to MSDN -- or anyone else, for that matter -- is also vulnerable to security breaches. Do you have evidence to suggest that Passport is less secure than MSDN's user database?
It seems to me that once you've given your data to one part of Microsoft, you've given it to all of Microsoft, and the fact that they're copying it from one database to another isn't that big a deal.
None of which is meant to defend their apparent goal of taking over the world's authentication services, which I do have a big problem with -- but although I admit I rolled my eyes when I got my copy of the MSDN E-mail this morning, I concluded that this isn't really a significant development.
There were plenty of crappy derivative games in the C64 days too. It's like old movies: we only remember the memorable ones, and after a while we forget Sturgeon's Law isn't a new phenomenon.
I remember at least seven or eight lousy "Defender" clones, endless streams of text adventures about as logical and coherent as today's click-on-the-surreal-picture equivalents, and I couldn't begin to count the run-around-the-maze-collecting-things games from that era.
It's not a C64, but I recently dug out my old Atari 800 and several big boxes of floppy disks full of pirated software (a habit I've long since given up, I hasten to add). On about one out of three disks there was a game I wanted to go back and play again; the rest made me think, "Oh yeah, I remember trying that piece of junk one or two times."
Slashdot Mirror
It baffles me that the author of the article could know about Interbase and yet seemingly not even look at it in the course of researching the article.
Of course, we'll see if they get to review ATI's next new card, but it's clear that the so-called catch-22 isn't universal.
Copyright law and patent law are two different beasts, though they both cover intellectual property. In the case of patent law -- which, keep in mind, also covers physical inventions -- there are economically significant players who stand to lose a lot if patents are extended indefinitely. It's not like copyright law, where the main opponents to extension are librarians, academics, and well-meaning citizens. There would arguably be well-funded corporate opposition to extending patent terms. (I agree there'd also be well-funded support for the idea, but it's not as grossly one-sided as is the case with copyright law.)
US patent terms have changed a lot less frequently than copyright terms, historically speaking. They changed from 14 to 17 years in 1861. They stayed that length for over a hundred years, until a 1994 amendment passed to implement the Uruguay Round of the GATT treaty, which changed the term to 20 years from application date.
There have been plenty of big companies with valuable patents over the years. To my knowledge there has never been a concerted lobbying effort to get patent terms extended; if there has been, it hasn't worked. One reason is that it's much easier for a legislator who isn't versed in IP law to see the social benefit of patent expiration than of copyright expiration. From lug nuts to tumbler locks, you can point to any number of real-world examples and easily get a lawmaker to agree that it would have made no sense for one company to own those technologies till the end of time. It's a much more abstract, fuzzy argument to say that it's good for society if Disney stops owning Mickey Mouse. (Which, I hasten to add, I think it is.)
In fact, we're already seeing it in a few cases. RSA's public-key patents are an obvious example. Three years ago a lot of people were livid about the restrictions those patents placed on developing crypto software. Now the patents are over and we have software like OpenSSL which are better than the commercial alternatives. (Granted, development of much of this software started long before the RSA patents expired, but the point is, nobody is worried about these supposedly horrible, disastrous patents any more.)
Obviously 20 years is a long time in the software biz, but the point is, it's not forever, and the situation will improve over time even if the bad guys win.
Stargate is just below my threshold -- not bad and I catch it pretty often, but I feel like I watch enough TV as it is so I don't follow it religiously.
To my thinking, the power and flexibility I'd want in a mobile device is pretty much proportional to the richness of interaction I can have with it. Cracking the rich-interaction problem on small mobile devices, it seems to me, is going to be a much tougher job than cramming in software and CPU power that used to only be available on the desktop.
Now, the interesting thing is what I do once I've decided to filter the mail. Since my rules catch legitimate mail, I don't just throw it away. I wrote a small collection of Perl scripts (which I'll release to the world someday soon, but they need documentation) that maintain a whitelist of sender addresses.
If a filtered message is from an address that's marked valid, it's delivered. If it's from an address that's marked invalid, it's discarded. If it's from an unknown address, the message is put in a holding area and an autoreply is sent back to the sender from a magic address asking them to reply in order to validate themselves.
The magic address is unique per filtered message -- it uses qmail's address extension mechanism -- and mail to the magic address never gets delivered to me, so I don't care if it gets added to spam lists. The Perl script behind the magic address does a quick check to make sure it's not processing a bounce, then marks the sender of the original message as valid and delivers the original message (or messages if more than one arrived while awaiting validation).
Held messages are cleaned out by a cron job when they get too old.
This is sort of similar in concept to the password mechanism of SpamBouncer or (a closer cousin) SpamCop's whitelist feature, but it doesn't require senders to retransmit their messages, which I always thought was pretty annoying to ask people to do since not everyone saves their outgoing mail. Granted, asking them to do anything is kind of annoying, but at least this is less so since they can just hit "reply" and "send".
This setup is cool because it allows friends to Bcc me on stuff without my "I must be listed as a recipient" rule trashing their messages, even if they've just switched E-mail addresses. It is admittedly based on the assumption that spammers don't read replies to their mail and/or wouldn't go to the effort of unlocking themselves; I have yet to see a spammer do that, and given the economics of spamming I think that'll be a safe assumption for the foreseeable future, unless this approach gets so popular that spammers start writing automated unlock bots!
Let me give you one of my questions from the story in question, which alone eliminated a good 40% of the candidates.
"In a SELECT statement, what's the difference between the WHERE clause and the HAVING clause?"
For those reading this who aren't SQL experts, that's vaguely equivalent to asking a Perl programmer what the difference between an array and a hash is. Yeah, it's possible to use Perl for years without knowing that, but it's a fundamental enough part of the language that if you can't answer that question you shouldn't claim to be an expert, and further, if you can't answer it and you do claim to be an expert, I become very nervous about how accurately and truthfully you've gauged the rest of the abilities you list on your resume.
I've asked interview questions I don't know the answer to, but never without clearly stating that fact in advance, and those questions were always high-level algorithm design, "see how they approach the problem" types of things. For example, at one point several years ago I was playing in my spare time with writing a program to lay out crossword puzzles (not as simple a problem to solve efficiently as it might intuitively seem) and I asked lots of candidates how they'd approach that. I was mostly looking to see what ideas they came up with and discarded as they thought the problem over. It was not important which specific approaches they tried; I wanted to see how they arrived at those approaches.
When I need to ask specific design questions, I always ask the candidate to solve some problem that I or one of my colleagues have just recently solved ourselves, again so I can see how their thought processes compare to our own. Which, BTW, also provides insight into how well a person will mesh with the rest of the team, though it's not the major way I judge that.
I've never worked for a company where I was expected to let HR grade someone's technical skills. That sounds like a recipe for disaster. You can't ask anything beyond extremely rudimentary questions and expect an HR person to make sense of the answers, especially since, as you say, looking at the person's thought processes is critical.
Or maybe it's just me.
I'll take one example. We had a data warehouse (mostly a big Oracle PL/SQL application). The engineer who designed and implemented the original code left the company, and I was tasked with hiring his replacement since there were some pretty substantial architectural changes we needed to make.
Now, there are a lot of database people out there. A lot. I looked at more resumes than I can count. My ear was sore from phone interviews. Thing was, just about everyone I talked to fell into one of two categories:
I looked and looked. The executive staff got really antsy and started leaning on me to do what the article suggests, just hire someone to get the work going, even if they weren't perfect for the job. I resisted for a while but finally caved in.
The contractor we brought in -- one of the better ones I'd interviewed, though I hadn't liked him well enough to want to hire him -- did a decent job of talking to the right people, gathering requirements, and getting himself acquainted with the layout of the code. But then he started to submit his own code, and man, what a disaster. I wasted weeks correcting his mistakes. Finally I fired him and went back to my original search.
The specifics of the story here aren't important. The point is that it doesn't take many times being burned by the "hire any bum off the street, just fill this technical position" attitude before you develop a very healthy caution about hiring the wrong person. I've seen it happen at other companies and I think it's a universal truth: hiring the wrong person for a job can leave you in a much worse position than hiring nobody at all. Not least because you think you have the position filled, so you stop looking for a while.
Experienced managers know this, so they put themselves through the "there's nobody out there!" routine when the job market is tight. It sucks massively, but it sucks less than the alternative.
(How did the story end? We found an H1-B person who fit the bill perfectly. Then the government took so long to process his paperwork -- months -- that by the time it came through, he'd gotten cold feet. Ugh! Happily by that time I'd moved to a different group.)
For those who don't know about it, I'll expand a little bit: DScaler and a TV capture card takes input from a regular NTSC video source (cable TV, VCR, TiVo, DirecTV receiver, whatever) and applies a bunch of video processing algorithms to remove the visible artifacts caused by the fact that NTSC is an interlaced video standard. It uses your PC's video card to scale the image up to whatever resolution you like. It'll even do fancy tricks like displaying film material at an even frame rate if your refresh rate is a multiple of 24Hz (the frame rate used for most film material), or automatically detecting letterboxed movies and expanding them to fill a widescreen display. Lots of people are using it in place of expensive standalone video scalers to display analog TV on their HDTV sets. If you have a clean source signal you can get a picture that looks nearly as good as standard-definition digital TV. Not HD by any stretch, but it still looks quite good.
Best of all, it's GPLed.
The main place where it's discussed, and where all the developers hang out (and a great place to discuss using PCs for watching TV and movies) is AVS Forum's Home Theater Computers message board.
The protests are at least getting a little bit of mainstream press coverage, which I hope will at least make the broader public aware that there's an issue here. Without the support of a bigger slice of the voting public than Slashdot's readership, there's no way enough pressure will be brought to bear to get the DMCA repealed. (Though of course court challenges are also a possibility.)
That said, for those of us in places like the San Francisco Bay Area, which has a large number of digital stations, DTV and HDTV are just lovely. On a clear day, my rooftop antenna picks up six or seven digital stations. The picture quality is stunning even on the standard definition stations, much crisper than the clearest cable channels and most DirecTV channels. And HD shows look better than the picture at the local movie theaters. The picture has yet to fail to elicit a "wow" when I've shown it to people.
And the cool thing is, it's on my computer using an ATSC tuner card which means I can record the digital signal to my hard disk for later viewing - not as slick as a TiVo, but adequate. (And before you ask why anyone would watch HDTV on a 17" monitor, the monitor on that PC is one of these, more or less, less expensive than a new HDTV if you buy it used.)
I do wish the prices would come down on more traditional HDTV sets and that they'd get the integration issues straightened out so a separate settop box wasn't required. Better market penetration will equal more incentive for the networks to produce more HD shows. But if you're willing to actually learn about what you're buying, the technology is out there and working.
You'd burn a copy of the CD without the protection for use in your car.
On the other hand, in many ways it's also the dumb idea behind Ultima Online, EverQuest, Asheron's Call, and other online games, some of which are making a tidy profit. By and large those games are pretty pointless without the other players. I think this is proof that people are willing to pay for the privilege of interacting with strangers, especially if the interaction is of a sort they can't easily get elsewhere.
The problem with Napster's pay-to-contribute model is that the contributions are, for all practical purposes, anonymous. Nobody will ever say "Oh, there's Joe, his ripped MP3s are the best!" I doubt one Napster user in 1000 paid the least bit of attention to who they were downloading from.
Throw in the power of ego gratification, and getting users to pay to create your content for you isn't quite so impossible.
Tea, Earl Grey, COLD.
As a mail admin myself, I've gotten my share of bogus SpamCop reports, so I can sympathize with what you're saying, but I think SpamCop's good outweighs its bad.
Because that's treating the symptom, and we'd rather get rid of the disease. If we can eliminate or greatly reduce spam with our tinkering and complaining -- and I know my tinkering has gotten rid of the majority of the spam I used to get, though I still get too much -- then we save that much time every day for the rest of our lives.
Because spam is about more than the two seconds it takes to recognize and delete the mail; it's about the irritation we feel every time we have to do it. Not a major irritation in isolation, but any minor irritation becomes a major one if repeated frequently enough.
And because we feel it's wrong to spam, plain and simple, and speaking out against what one feels is wrong is, we feel, a right and a duty in a free society.
I use 100Mbps on my home network. 10Mbps doesn't cut it if you're doing any work with video files, for example, or doing big builds over the net. What actually prompted me to upgrade was how long it was taking me to copy my ripped MP3 files to my fileserver, if you must know.
Just to name a few. And of course you typically want a low-RPM hard disk (though the new quiet Seagate drive sounds promising), low-rotation CD-ROM drive (use a drive with Zen's TrueX multibeam technology and you'll still have fast reads), etc.
Apple's iMac Cube is proof that it's possible to build a computer with no fans whatsoever. Hopefully it's a harbinger of things to come.
Day after day, making tomorrow seem like yesterday!
Right-drag on a link with a down-and-up motion and it opens the new page *below* the current one in the list of windows so you can keep reading whatever page you're on; when the new page finishes loading its entry in the window list changes color.
This is great for Slashdot; by the time I'm finished browsing the top few comments, I can switch (again, using nothing but mouse gestures) to the story being discussed, no twiddling my thumbs waiting for some Slashdotted page to load, or worse, switching back and forth among three windows checking to see if two of them have finished loading yet.
Your point about multiple monitors is a good one, though. I only have the one monitor so it hasn't been an issue for me.
Opera doesn't deal well with things like drop-down menus implemented using CSS and JavaScript, and it's a bit crash-prone (though less so than Mozilla), but it's my browser of choice by a long shot.
Both of which, it seems to me, are kind of shaky complaints. If the current terms of use are troubling, that's one thing, but the fact that the bad bits have been removed means that the bad bits aren't there to complain about any more. Yeah, they could add the bad bits again in the future, but no more so than any company could modify their terms of use after the fact (e.g. Microsoft could modify the MSDN site terms of use in exactly the same way).
And while I certainly agree that Passport is a much juicier cracking target than the MSDN user database, the fact is that any personal data you supply to MSDN -- or anyone else, for that matter -- is also vulnerable to security breaches. Do you have evidence to suggest that Passport is less secure than MSDN's user database?
It seems to me that once you've given your data to one part of Microsoft, you've given it to all of Microsoft, and the fact that they're copying it from one database to another isn't that big a deal.
None of which is meant to defend their apparent goal of taking over the world's authentication services, which I do have a big problem with -- but although I admit I rolled my eyes when I got my copy of the MSDN E-mail this morning, I concluded that this isn't really a significant development.
People had more aesthetic sense back in the day?
I remember at least seven or eight lousy "Defender" clones, endless streams of text adventures about as logical and coherent as today's click-on-the-surreal-picture equivalents, and I couldn't begin to count the run-around-the-maze-collecting-things games from that era.
It's not a C64, but I recently dug out my old Atari 800 and several big boxes of floppy disks full of pirated software (a habit I've long since given up, I hasten to add). On about one out of three disks there was a game I wanted to go back and play again; the rest made me think, "Oh yeah, I remember trying that piece of junk one or two times."