As a Canadian currently on an H1-B, I can attest to the lowered cost of education. But really, is it my fault that my country decided to invest in education and subsidise it, rather than charging me $10,000+ per year for tuition? (For the record, I paid about $3,000-$3,500 per year for tuition, depending on the year, at a university ranked in the middle by Maclean's national university surveys).
When I first came into the United States, I filled a position that the employer had open for over six months and no Americans wanted to fill it. Now I'm on my sixth year of my H1-B, and am currently in the process of a Green Card application. My H1-B extension (for the purposes of getting the Green Card) is being processed, which will let me stay here longer.
While I would like to get a Green Card and become a permanent resident, there is no way in hell that I will ever become a U.S. citizen. As a (completely legal!) foreigner, I am treated like crap by your government and (some of your) people, and I have no faith that things will get any better. The current rules in place basically tie me to my current employer until I either abandon my Green Card and leave, or through some miracle the government goes through the backlog (which will take eight years to do, and there's no signs of it getting any better).
If you want to get rid of indentured servitude, then you don't need to get rid of guest workers and H1 programs, you need to improve your internal processes so that we can move from employer to employer more easily and not have to be in paranoid fear of the men in black walking us to the border because we made a typo on a piece of paperwork.
You're restricted to mpeg2 so disk usage is high and downloaded videos have to be transcoded to play, you can't play games or have pretty music visualizations, and sound is a minor configuration headaches.
Not true, if you have a PC with half-decent horsepower.
The latest version of the ivtv driver (0.4.1, although it was supported since the late 0.3.x series) when paired with the latest version of the Xdriver allows for Xv playback. The decoding is done in software via MythTV (or whatever other app you use) using Xv, so you're not limited to MPEG-2 playback. You can use xmame fullscreen (use -vidmod 1), with the exception of games that are higher than 720x480 (but you can work around that with -noperfect-yuv, although you might not have optimal picture quality).
The best part about using the PVR-350 output is that it properly handles interlaced and deinterlaced content. Although MythTV does have Bob deinterlacing, it really messes up the OSD (On Screen Display), making it flicker. There are no such problems with the PVR-350 output.
It seems that using the MPEG-2 decoder causes the odd lockup, but I haven't seen any lockups with the Xv driver. And there's no comparison in picture quality over my GeForce 5200FX.
Since I've been using my PVR-350, I have been living without MAME, which does kind of suck. But once X11R6.9 or X11R7 are out in RPMs for Fedora Core 4, I'm looking to do a dual-head setup with the GeForce 5200FX for xmame, and the PVR-350 for MythTV. I'm also looking forward to the ivtv team fixing the over/underscan on the PVR-350 - there should be a register in the TV-Out chip (the SAA?) that allows for scaling. I took a quick look at the datasheet and didn't see anything, but might look at it again.
Even if you do get a visa for 3 years, and extend it for another 3, after 6 years you MUST leave the USA.
IANAL, but I have spoken to an immigration attorney about this subject (and my company has this attorney on retainer as they're sponsoring me). I am an H1 holder currently on the second three years (it expires late next year). As long as you have a valid status, you can extend your H1 (or even a TN, although it's trickier) almost indefinitely. Valid status would include applying for a Green Card and "being in the queue". You're now allowed at a minimum, H1 extensions for one year as long as your application is being processed, although typically they're handing out three year extensions.
A word on trying to get a Green Card: I beat the deadline of the new "PERM" program for Green Cards,and as such, I'm under the old system (RIR). Because of the huge number of applications that went in (some 350,000), it will likely be at least another year before they even begin to unseal my file and start processing it (I, along with many others haven't even received an official receipt notice of my application). Then, with only about 140,000 immigrant visa numbers given our per year, my application will likely take an extra four years after they have approved it (which could take another three years), because they're giving out IV numbers to applications from 2001 (and even worse, each person attached to a Green Card application gets one number, so a family of four takes four IVs). So, with any luck, I'll have a Green Card somewhere around 2010-2012. By then, I'll either be married to an American citizen, or back in Canada. The new system (PERM) has problems of it's own, they're (seemingly randomly) rejecting applications and they're not even set to process the ones that they're not rejecting. It's a complete mess.
I live in the Valley, and anything less than $60K/year is gettng pretty difficult to live (a decent lifestyle) on.
I started out in 1999 at $45K/year in the Valley. It was very rough, I had enough for my apartment in a crappy (pun not intended) area of town (i.e. walking around human feces and homeless people on the sidewalks) and a bus pass, although I did tend to walk to work a fair amount to save money. After the bare necessities, I did have some money left over for some luxuries, like cable TV and DSL. But, I didn't have a car - and living in the Valley without a car makes a lot of things difficult, like grocery shopping. Not a whole lot of money left over for toys, and forget about supporting a family on that. Luckily my student loans weren't that bad, so I could afford to make payments on them.
I can tell you the only reasons why I survived on that salary: 1. I had very little furniture in my cramped studio apartment. My TV was a Commodore 1702 monitor (12 or 13", I think) with a cable converter. My drawers were baskets that held my clothes 2. I didn't have a car. Therefore, no insurance payments or gas. 3. I had lived on my own for the previous six years (four of them in a dorm room), and was used to having nothing (in the way of personal belongings) except for my computer, toiletries, and clothes. 4. (This is the most important) I had little debt, so I was able to make the minimum payments and keep people off of my back. But honestly, that's not much of a life to live, and almost everybody wants better than that.
I'm grateful for the experiences that I had and do have upcoming with the company I'm currently working, but I wouldn't be able to do it now in 2005 on only $45K per year (which is what I'm guessing you'd be offering - $45-$50K per year). It's just too expensive to live in Silicon Valley on such a small salary.
If you really want talented people, then I think you'll have to pony up the $60K per year. After getting laid off from that $45K/year job (in 2001), I was asking for almost double, and easily got it. Back then, people were willing to work for promises of better pay, now people don't buy into that hype and want the money up front. As far as I'm concerned, stock options and promises are worth the paper they're printed on, and nothing more, until I actually get cash in my bank account from them.
I feel your pain. I too am in a similar situation.
The company that I work for, although we're not a massive player in the media industry, we're trying to break into it and do some revolutionary things. (I cannot comment on these naturally.)
That said, I know that management must feel a bit uneasy with me. They hired me because they know me from my work at another company. They feel that I do a very good job, and that the software that I write is of high quality. However, they also know how I feel about DRM, and since I'll eventually be working on some sort of DRM for the system, well, they can't be too happy with the conflict between my beliefs and the business plan.
I'd love to see things like CableCard cracked, but my employment hinges on things like CableCard being successful in the market and content producers getting paid for their content.
It may be a bit expensive, and it might only work with Linspire, but it's completely legal.
The RIAA/MPAA doesn't give two shits about Open Source or Free Software users. They want the content played on their terms (which is pay per use for infinity). The Open Source/Free Software users are actually a thorn in the side of some companies due to "free" implementations. That MPEG-2 decoder xine (and others) use is using is violating patents. The first "free" version of MPEG-4 was a hacked up Microsoft codec. Current open source MPEG-4 implentations violate patents. LAME violates patents surrounded by MP3s, Etc, etc.
When will Linux users get CableCard? If the "powers that be" have their way, never.
Please understand that OpenCable is run by CableLabs, a very corrupt, yet inept organization, which is controlled by the major cable conglomerates (mainly Comcast) and the big players in the video/data space (Motorola and Scientific Atlanta). The cable operators are allowing CableCard only in the hopes that they can use it as a way of leveraging the existing cable headend distributors (Scientific Atlanta and Motorola) into lowering their prices and (potentially) becoming interoperable. (Currently in the cable market, once you pick a headend supplier, you're locked in with them, because they have incompatibile proprietary encryption schemes). Believe me, they really don't care about the end customer, as long as the customer doesn't switch to IPTV over DSL (which is a big threat now).
If enough people switch to Linux on desktops and set top boxes, market forces will drive alternate solutions.
My friends tell me that I'm very pessimistic - I prefer to call myself a realist. Nonetheless, Linux is starting to drive set top boxes because it's cheap. And the DRM is done via a SmartCard that plugs into a reader that is controlled by Linux. But, the reader is controlled by a binary-only module (although the set top box designers have the source code), so the end user will never know. And encryption is handled in the hardware, the bits actually travel across the buses encrypted, so it would take a lot of expensive equipment to reverse engineer it.
Linux on desktops, well, there's MythTV. But, if the developers can't figure out CableCard without violating the DMCA, then it won't happen. Some developers might figure something out, but it will never be part of the main codebase. Heck, they won't even allow talk about Torrentocracy on the myth-users mailing list because of the fear that MythTV will be seen as a bad program for allowing you to download TV shows.
All in all, it's a sad state of affairs, but unfortunately, that's the way it's going. At least in Europe, they have some idea on open standards, with things like DVB.
Myself though, I agree with you. I'm sticking with SDTV because I don't feel like paying a lot of money for a TV that will either try to lock me out of the content I'm paying for, or will be made obsolete with new DRM restrictions. I'm staying away from Digital Cable for the same reasons. But, unfortunately, most people just don't care.
-- Joe
Re:XBOX series *LACKS* important DC features
on
XBOX 360=Dreamcast 2.0?
·
· Score: 2, Interesting
My, my, we wouldn't be a disgruntled Dreamcast fan who waited for Half-Life, would we?:) (Or am I being trolled? I'm not really sure!)
Seriously though, them using Windows CE for Dreamcast played a major role. They couldn't keep a decent frame rate (which probably came from the combination of C++/WinCE/DirectX), and they were having trouble with sound. (The sound one I recall in particular because they wanted to do multiple effects on a playing sound, and DirectSound wasn't capable of it, because it was running an older sound driver that didn't support it. I think that the Sega sound driver supported it though, I just don't remember that far back).
Then there's the usual politics, return on investment questions, and the lead developer was apparently having personal problems which led him to taking some time off from work.
I guess that everything combined just caused them to drop the title. It clearly wasn't because of the Dreamcast hardware - the port of Quake III was quite good, and as I recall, it was one programmer who did everything.
-- Joe
Re:XBOX series *LACKS* important DC features
on
XBOX 360=Dreamcast 2.0?
·
· Score: 5, Interesting
I started at Sega just before the launch of the Dreamcast, so I don't know the entire story (or even if this one is completely true). Nonetheless, this is what I heard: The team at SOJ (Sega of Japan) didn't want to make their own OS for the Dreamcast. They were uneasy about it, but chose Windows CE, as I guess they figured that it was similar enough to Win32. The fact that Microsoft was able to demo Internet Explorer on the Dreamcast was probably a big bonus, given Sega's desire to make the Dreamcast a big thing on the Internet.
However, as time went on, SOJ noticed that Windows CE was big and bloated and full of bugs. There was developer backlash. Sega's own software development teams (AM) needed something better, especially if they were to make full power of the Dreamcast (there was an arcade system that was basically a Dreamcast on steroids, although the name escapes me now). Thus, Sega of Japan started to develop their own low-level operating system for the Dreamcast. By the time that this happened though, the contract was in place with Microsoft for Windows CE - part of the contract was that Sega had to make WinCE available to all developers and stamp the logo on the unit.
My own observations: Most of the developers in the United States and Europe used Sega's OS - it just provided the low-level functionality that tbey were used to. The developers who were using WinCE usually had an existing code base that ran on Win32 on the PC, and they were looking for a quick port. For games that weren't really intensive, this worked fine, but for some others (I saw early versions of Half-Life on the DC, when it was already delayed by at least six months, and let me tell you, it had *major* problems+) it was a disaster. What I would tell developers who were asking me if WinCE was worth using: "It will get you a solid 15 FPS, and if you're willing to optimize your rendering code with some assembly (to make use of the SH4's vector functions), you should be able to get 30 FPS. You'll get your game up and running faster, but you'll spend more time optimizing it for speed."
Version 1 of WinCE for Dreamcast was pretty buggy, version 2 fixed a lot of things, but even once they came out with the final version (I think 2.1), there were still lots of bugs. Developers were asking me if Microsoft would release WinCE 3, or at least fix some of the bugs. I tried to get the source for WinCE 2.1 for DC (so that I could try to at least maintain it myself for the developers) and got nowhere. Mind you, this was only a few months from Sega canning the Dreamcast anyway.
Microsoft did what they could to get developers to use WinCE on the Dreamcast. They'd send out promotional material to every new (and existing) Dreamcast developer which included a T-shirt, a Leatherman tool and of course, the WinCE SDK CD. We got a lot of thanks for the free tool and a T-shirt, while the CDs went into the garbage.
-- Joe
+ That's not to say that the sole decision to cancel Half-Life for the Dreamcast was because of WinCE performance. There were also some other issues outside of that, which I shall not discuss, but WinCE was a big factor.
No, I mean Windows Autorun, which determines when a CD has been inserted.
My guess is that (and I'm not familiar with Windows APIs, so I don't know if this is even possible) iTunes registers an event handler for when a new CD is inserted, and that triggers it to do ripping.
I disable Windows Autorun because I happen to have a CD that won't let you listen to it's content without installing a stupid program (or so I think - Exact Audio Copy will see the tracks, but Windows Explorer insists on opening only the data portion, and WinAmp can't open the tracks either), and I'd rather not install that program. (BTW, no, I did not buy this CD, it was a gift from a relative of mine who isn't too technically inclined. And even worse, they don't say that the CD is copy protected, they just say that it's "web-enabled", or something like that. But, there is a lack of the CD Audio logos in the packaging/jewel case). Also, Autorun just annoys me.
The help in iTunes (or, at least a dialog box that popped up) says that with Autorun disabled, I can signal to it that a new CD has been put in - I need to click outside of the iTunes window, and then click in it again. But that doesn't seem to work. Again, not an issue, since I'm not using iTunes anyway (except to sync up video and pictures the odd time).
Try Anapod Explorer from Red Chair Software instead of iTunes (http://www.redchairsoftware.com/anapod/). I find it works quite well on my shiny new iPod Video (for music only, but I'm sure that they'll have video support in due time).
iTunes has huge problems, but the one that bugs me the most is that I can't seem to get it to rip CDs with Autorun disabled. But, now that I'm using Anapod Explorer, it's a non-issue, since I'm re-ripping my CD collection into FLAC, and Anapod will convert it to WAV or MP3 on the fly before uploading it to the iPod.
It's a quality program, and I've been using it since 1992 on my Creative Nomad Jukeboxes.
In Canada, there's certainly no shortage of tech workers, but there is a shortage of tech jobs (though, I hear that is improving now with a bunch of tech companies in and around the Waterloo, Ontario area).
The Canadian government is afraid of "brain drain", and rightfully so - a lot of their bright technical people are leaving for the U.S., because they can't find jobs in Canada in their chosen field. I left Canada six years ago to go to California, where I'm working as an embedded systems programmer. My first employer came to me - this after some six months of job hunting within Canada, before and after graduation. I had never given thought to leaving Canada until I got that offer, and it was certainly better than working in a computer store, or working phone lines.
In fact, two of my friends from high school also came down to California (although both are back in Canada now), and I've met lots of Canadians down here. They all say the same thing - no (non-minimum wage) jobs in Canada for them, so they came down here.
I live in Northern California, but it wouldn't surprise me if we have the same provider. My services are provided by CCI (the full name eludes me right now), which is based out of Riverside, CA.
I've been using it for a couple of weeks now, and it's not so bad.
I'm actually in the same situation, I moved into a new apartment complex where they do "repackaged DirecTV" also. It's actually not that bad, but the biggest problem that I have is that I never know whether the glitches in the video are because of the signal, or my PVR-350/PVR-500. That and Zap2It (used by my MythTV machine) has incorrect listings for some of the channels (because they're on an EST feed, but Zap2It thinks that they're PST channels).
The only problems that you'll really have are when the dish can't pick up the signal, and possibly when the content is shown (like I said, most of my channels are on an Eastern feed, but I'm in California). But, I just record the shows that I want to watch, and it works out well enough. And, as a bonus, I get my cable TV, phone service, and internet for cheaper than Comcast cable and Internet services.
The underperforming people would definitely try to get into the union (I'm sorry that I didn't explictly state that). They would try to unionize because then they collectively would be able to force the employer not to get rid of them. The above-average performers would be dead-set against the union.
Why would the lower-end quit the union? That union is the only way they'd have their jobs! While the above-average people have a proven track records, thus making them attractive to other employers, the below-average do not.
So, let's say that a tech union manages to materialize at a job. Here's what I predict would happen: - The above-average performers would quit the union (which likely would mean changing jobs), because: 1. They'd be sick of their underperforming co-workers floating off of them. 2. Unions actually sometimes set maximum caps for productivity, in order to make their underachievers not look so bad. - The lower performance employees would stay in the union, because the union guarantees their jobs. - Employers cannot refuse employees right to unionize... So they'd probably close down shop, and reopen. But, they'd know who was part of the union, and refuse to hire those people (not because they were union, but because they were underachievers). - Meanwhile, the more productive employees who quit the union have moved on to another job where the employers prune the deadwood, so that there's no incentive to unionize.
There are several good companies out there, and I work for one. We're a small company, but all of us know what we have to do, and we get it done. We've never even thought of unionizing because we're treated very well, and get paid very good wages.
After all, that is theoretically what unions are supposed to offer employers, higher quality employees. Most unions do not do that. Most unions are leeches, taking their members' dues and not helping out the members. They also cripple business. And most of the better tech people know this and stay well clear of them.
I would certainly hope not (although I wouldn't like to be the person to test that), mainly because if our employers don't owe us two weeks notice, then why should we?
That said, I have worked in companies where they have always provided advance notice of layoffs, and I have always given two weeks notice, if nothing else, to not burn bridges (and in some cases, even helped them out after those two weeks with my new employers' permission).
Do you honestly think that underperforming employees who become part of a union would agree to that? No, in fact, I'd wager that many of the employees who would benefit from a tech union would be the poorly performing employees.
In fact, I'd even say that the reason we don't have tech unions are because of the better performing employees who refuse to unionize for just that reason.
Also, the rules aren't necessarily put in place by the employees, they're put in place by whoever controls the union. This would not likely be tech workers, but instead another group of people (i.e. the local union), which wouldn't know the first thing about what the "lower rung" should be.
I agree wholeheartedly with everything that you said. I specifically chose not to mention patents though, because there are things that you could learn, that while not necessarily patentable, do seperate your company from others. (Perhaps an efficient algorithm for filtering packets, which although is blatently obvious, your competitors haven't figured it out, or felt the need to figure it out due to lack of competetion.)
I disagree with the, "All you know, and all you create while you work with us belong to us" contracts, and anytime I see language like that, it puts up a red flag. I immediately try to negotiate that. (In fact, my previous employer had something like that in the contract, I explained that California State Law doesn't allow for that, and I wanted it removed. I also got them to put in a clause explicitly stating that anything I did in my own time belonged to me, just so that there was no confusion.)
I just have a policy of not leaving it up to the courts. If something in a contract is unclear, it needs to be made clear.
As to your point about rewards for big ideas, well, I can't say that I've been in that situation yet, so I can't really comment either way.
Like I said in another post, I'm lucky - I've only had one employer who has treated me poorly in my professional career (I've had some who treated me poorly before however).
I would never participate in a tech union, because given my experience over the last six years (four different companies, but smaller ones), I know that I could do better.
Unions served a very useful role when they came out, but I've talked to (former) unionized workers recently, and their common conclusion is that unions protect the weakest people best - those who can't fulfill their job responsilibities very well (i.e. meet the quota, and a reasonable one at that).
I'd rather perserve my job by showing management that I can work better than those people. It's worked for me very well thus far in my career, as I've increasingly been given more important roles and pay raises, while the lesser skilled people were shown the door.
An example of the crap I've seen: One software developer who spent three weeks on an issue, and every time when asked why it wasn't fixed, would say, "Hardware problem. I can't fix it". The boss would re-assign the bug to me, and I'd have it fixed in two days (without any of their bugtracking notes, since they never logged any in the system, so I came in cold) - definitely not a hardware problem, but an unskilled person.
Then again, as I've said, I've worked at smaller companies (at most 20 technical people in my group), and my employers were good enough to give me a chance to show my worth (and reward me for it). I'll admit that I've been very lucky in that regard. I also don't put in any more than 45 hours in a given week (unless an immediate problem comes up), and my manager is perfectly fine with that.
As rediculous as that sounds, that's not a bad idea.
If you can't understand the terms of any contract, you should definitely consult a lawyer before signing it. Also, I'd be wary of working for such a company anyway - a company that has to spell things out in such a manner (beit explicit or obfuscated) likely wouldn't be a company I'd want to work for anyway. I've found that I work better when I have fewer restrictions like that.
On the other hand, I dispise the fact that things have gotten so bad in employment.
No, this is why people should be very diligent in reading the fine print on contracts.
If you can't abide by the terms of the contract, then you either negotiate the contract, or don't take the job, it's that simple. If people wouldn't accept the jobs, then companies like Microsoft would have to treat their employees better.
Sidenote 1: From the time I spent at the Microsoft campus in Richmond (one week in late December 2000), I got the impression that full-time employees are treated quite well. Everybody has their own office (most people suffer in cubeville), and they are encouraged to make their office very personal - those who are musically inclined actually have their instruments there, for instance (well, smaller instruments). Of course, this is part of then encouraging them to work longer hours, but it wasn't forced to the best of my knowledge. Just avoid the cafeteria food, I got pretty sick off of some of it.
Sidenote 2: I wouldn't work for Microsoft, not that I couldn't have at least gotten an interview. I wouldn't work for them, just because many of the employees, while great people, are just corporate zombies - you flick a switch, and they say, "I can't understand why the government is out to get us, we've done nothing wrong!"
But that's totally unrelated to the point of my comment.
If you meant non-compete, then I agree with what you said. If you did mean non-disclosure, then I disagree.
In the case of a non-compete, since the company would be preventing me from working for another company in the field, I definitely should get a golden parachute. Should I obtain a position at a company that is not considered a competitor, the terms of the contract can be renegotiated.
Non-disclosures are a different beast. I feel that you shouldn't be able to use any proprietary information that you obtain at one company at another (for instance, you develop an algorithm that completely revolutionizes search engines, you can't give that to a new employer). Generic knowledge on the other hand, is fine. Even in search engine technology, there's a lot of general knowledge.
I live in California, and the NDAs that I have signed basically state that I won't take any company secrets with me to a new employer. General knowledge isn't considered company secrets by any extent, so I'm quite free to move (and move I have, I've changed companies four times in the last six years).
Most importantly, I read the paperwork before I sign it, and if I disagree with it, I negotiate. At the previous company I worked at, they didn't want to give me any vacation time for the first six months, and also wanted to only give me 10 days (the norm is 15). I explained to them that I was taking over 10 days of vacation time in three months, and if they didn't like that, they could look for somebody else (though not in those words). I still got the job, because they felt I was the best candidate.
To some extent that is true, but not always. The sticky issue is that of device drivers. Device drivers hook into the kernel, and their licensing as such has never really been clear. The general consensus seems to be that as long as you mark your module as non-GPL in the declaration, which excludes your use of some GPL'd code, your module is okay.
Userspace applications don't suffer from these restrictions - glibc is LGPLd, as is uClibc for these reasons - you don't have to abide by the GPL in order to have a C runtime library.
Where I see this going (as I stated in another post) is that whether Linux or BSD is used will likely depend on the hardware designers (companies like Broadcom, who make the reference designs, not companies like Linksys, D-Link or Netgear, who just base their work on reference designs).
The sky is not falling, I agree. We'll continue see substandard products from el-cheapo manufacturers no matter what underlying OS is used. I am going to go out on a limb and say that if FreeBSD can be used in these low-power, slower CPU, small RAM/flash footprint devices, and it performs as well as Linux, then the designers will do so. It gives them that much more protection against violations of the GPL (accidental or not).
I posted something to this effect on the original CNet article
I work in embedded development myself (previously video game consoles, then DOCSIS cable modems, now video equipment), so I've seen the shift from expensive proprietary systems (like vxWorks) to free (as in money) systems like Linux.
The proprietary systems typically have high up-front costs, along with a per-unit royalty, which inflates the cost of the devices. Linux allows for cheaper devices (whether or not the savings are passed to the customer remains to be seen), at a cost (complying with the GPL). This can be somewhat mitigated by making modules that are not licensed under the GPL.
BSD entering the space will provide some good competition for Linux. Whether newer designs switch to BSD will depend on the chipmakers (like Broadcom), as they are the people who usually write the drivers. Most devices nowadays are just the reference design hardware tweaked a bit with the reference software. So, whatever OS is used for the reference designs is what will be the dominant OS in the embedded space.
Only time will tell, but if FreeBSD can pull this off, they'll definitely gain some traction.
Nowadays because the margins on networking hardware are so small, Netgear, D-Link, et al basically take a reference design and slap a different web interface on it. There's no differentiation in the actual software that controls the wireless portions. Updates to the software come when the chipset vendor releases updates to their customers. Then to top it off, the hardware is manufactured in Taiwan with the cheapest possible components. (I worked for a Taiwanese company making DOCSIS cable modems, that's the way that they all operate).
However, the 2.4GHz band is absolutely awful. In my apartment complex, there's a half-dozen networks around me (802.11b/g), to the point where I was having interference issues with the other networks. My solution was to invest in 802.11a equipment, which is working well, although quite expensive (well, not really expensive, but twice the cost of your average 802.11g system).
I'm actually a fan of the Netgear client for my Netgear PCMCIA card. It works quite well after a Windows re-install, but the software definitely doesn't like having drivers/software around for about a half-dozen other cards around (testing compatibility with other cards).
Slashdot Mirror
As a Canadian currently on an H1-B, I can attest to the lowered cost of education. But really, is it my fault that my country decided to invest in education and subsidise it, rather than charging me $10,000+ per year for tuition? (For the record, I paid about $3,000-$3,500 per year for tuition, depending on the year, at a university ranked in the middle by Maclean's national university surveys).
When I first came into the United States, I filled a position that the employer had open for over six months and no Americans wanted to fill it. Now I'm on my sixth year of my H1-B, and am currently in the process of a Green Card application. My H1-B extension (for the purposes of getting the Green Card) is being processed, which will let me stay here longer.
While I would like to get a Green Card and become a permanent resident, there is no way in hell that I will ever become a U.S. citizen. As a (completely legal!) foreigner, I am treated like crap by your government and (some of your) people, and I have no faith that things will get any better. The current rules in place basically tie me to my current employer until I either abandon my Green Card and leave, or through some miracle the government goes through the backlog (which will take eight years to do, and there's no signs of it getting any better).
If you want to get rid of indentured servitude, then you don't need to get rid of guest workers and H1 programs, you need to improve your internal processes so that we can move from employer to employer more easily and not have to be in paranoid fear of the men in black walking us to the border because we made a typo on a piece of paperwork.
-- Joe
You're restricted to mpeg2 so disk usage is high and downloaded videos have to be transcoded to play, you can't play games or have pretty music visualizations, and sound is a minor configuration headaches.
Not true, if you have a PC with half-decent horsepower.
The latest version of the ivtv driver (0.4.1, although it was supported since the late 0.3.x series) when paired with the latest version of the Xdriver allows for Xv playback. The decoding is done in software via MythTV (or whatever other app you use) using Xv, so you're not limited to MPEG-2 playback. You can use xmame fullscreen (use -vidmod 1), with the exception of games that are higher than 720x480 (but you can work around that with -noperfect-yuv, although you might not have optimal picture quality).
The best part about using the PVR-350 output is that it properly handles interlaced and deinterlaced content. Although MythTV does have Bob deinterlacing, it really messes up the OSD (On Screen Display), making it flicker. There are no such problems with the PVR-350 output.
It seems that using the MPEG-2 decoder causes the odd lockup, but I haven't seen any lockups with the Xv driver. And there's no comparison in picture quality over my GeForce 5200FX.
Since I've been using my PVR-350, I have been living without MAME, which does kind of suck. But once X11R6.9 or X11R7 are out in RPMs for Fedora Core 4, I'm looking to do a dual-head setup with the GeForce 5200FX for xmame, and the PVR-350 for MythTV. I'm also looking forward to the ivtv team fixing the over/underscan on the PVR-350 - there should be a register in the TV-Out chip (the SAA?) that allows for scaling. I took a quick look at the datasheet and didn't see anything, but might look at it again.
-- Joe
Even if you do get a visa for 3 years, and extend it for another 3, after 6 years you MUST leave the USA.
IANAL, but I have spoken to an immigration attorney about this subject (and my company has this attorney on retainer as they're sponsoring me). I am an H1 holder currently on the second three years (it expires late next year). As long as you have a valid status, you can extend your H1 (or even a TN, although it's trickier) almost indefinitely. Valid status would include applying for a Green Card and "being in the queue". You're now allowed at a minimum, H1 extensions for one year as long as your application is being processed, although typically they're handing out three year extensions.
A word on trying to get a Green Card:
I beat the deadline of the new "PERM" program for Green Cards,and as such, I'm under the old system (RIR). Because of the huge number of applications that went in (some 350,000), it will likely be at least another year before they even begin to unseal my file and start processing it (I, along with many others haven't even received an official receipt notice of my application). Then, with only about 140,000 immigrant visa numbers given our per year, my application will likely take an extra four years after they have approved it (which could take another three years), because they're giving out IV numbers to applications from 2001 (and even worse, each person attached to a Green Card application gets one number, so a family of four takes four IVs). So, with any luck, I'll have a Green Card somewhere around 2010-2012. By then, I'll either be married to an American citizen, or back in Canada.
The new system (PERM) has problems of it's own, they're (seemingly randomly) rejecting applications and they're not even set to process the ones that they're not rejecting. It's a complete mess.
-- Joe
I live in the Valley, and anything less than $60K/year is gettng pretty difficult to live (a decent lifestyle) on.
I started out in 1999 at $45K/year in the Valley. It was very rough, I had enough for my apartment in a crappy (pun not intended) area of town (i.e. walking around human feces and homeless people on the sidewalks) and a bus pass, although I did tend to walk to work a fair amount to save money. After the bare necessities, I did have some money left over for some luxuries, like cable TV and DSL. But, I didn't have a car - and living in the Valley without a car makes a lot of things difficult, like grocery shopping. Not a whole lot of money left over for toys, and forget about supporting a family on that. Luckily my student loans weren't that bad, so I could afford to make payments on them.
I can tell you the only reasons why I survived on that salary:
1. I had very little furniture in my cramped studio apartment. My TV was a Commodore 1702 monitor (12 or 13", I think) with a cable converter. My drawers were baskets that held my clothes
2. I didn't have a car. Therefore, no insurance payments or gas.
3. I had lived on my own for the previous six years (four of them in a dorm room), and was used to having nothing (in the way of personal belongings) except for my computer, toiletries, and clothes.
4. (This is the most important) I had little debt, so I was able to make the minimum payments and keep people off of my back.
But honestly, that's not much of a life to live, and almost everybody wants better than that.
I'm grateful for the experiences that I had and do have upcoming with the company I'm currently working, but I wouldn't be able to do it now in 2005 on only $45K per year (which is what I'm guessing you'd be offering - $45-$50K per year). It's just too expensive to live in Silicon Valley on such a small salary.
If you really want talented people, then I think you'll have to pony up the $60K per year. After getting laid off from that $45K/year job (in 2001), I was asking for almost double, and easily got it. Back then, people were willing to work for promises of better pay, now people don't buy into that hype and want the money up front. As far as I'm concerned, stock options and promises are worth the paper they're printed on, and nothing more, until I actually get cash in my bank account from them.
-- Joe
I feel your pain. I too am in a similar situation.
The company that I work for, although we're not a massive player in the media industry, we're trying to break into it and do some revolutionary things. (I cannot comment on these naturally.)
That said, I know that management must feel a bit uneasy with me. They hired me because they know me from my work at another company. They feel that I do a very good job, and that the software that I write is of high quality. However, they also know how I feel about DRM, and since I'll eventually be working on some sort of DRM for the system, well, they can't be too happy with the conflict between my beliefs and the business plan.
I'd love to see things like CableCard cracked, but my employment hinges on things like CableCard being successful in the market and content producers getting paid for their content.
-- Joe
Example: there is still no legal solution for playing DVDs on Linux.p hp?product_id=11804
Not quite true: http://www.linspire.com/lindows_products_details.
It may be a bit expensive, and it might only work with Linspire, but it's completely legal.
The RIAA/MPAA doesn't give two shits about Open Source or Free Software users. They want the content played on their terms (which is pay per use for infinity). The Open Source/Free Software users are actually a thorn in the side of some companies due to "free" implementations. That MPEG-2 decoder xine (and others) use is using is violating patents. The first "free" version of MPEG-4 was a hacked up Microsoft codec. Current open source MPEG-4 implentations violate patents. LAME violates patents surrounded by MP3s, Etc, etc.
When will Linux users get CableCard? If the "powers that be" have their way, never.
Please understand that OpenCable is run by CableLabs, a very corrupt, yet inept organization, which is controlled by the major cable conglomerates (mainly Comcast) and the big players in the video/data space (Motorola and Scientific Atlanta). The cable operators are allowing CableCard only in the hopes that they can use it as a way of leveraging the existing cable headend distributors (Scientific Atlanta and Motorola) into lowering their prices and (potentially) becoming interoperable. (Currently in the cable market, once you pick a headend supplier, you're locked in with them, because they have incompatibile proprietary encryption schemes). Believe me, they really don't care about the end customer, as long as the customer doesn't switch to IPTV over DSL (which is a big threat now).
If enough people switch to Linux on desktops and set top boxes, market forces will drive alternate solutions.
My friends tell me that I'm very pessimistic - I prefer to call myself a realist. Nonetheless, Linux is starting to drive set top boxes because it's cheap. And the DRM is done via a SmartCard that plugs into a reader that is controlled by Linux. But, the reader is controlled by a binary-only module (although the set top box designers have the source code), so the end user will never know. And encryption is handled in the hardware, the bits actually travel across the buses encrypted, so it would take a lot of expensive equipment to reverse engineer it.
Linux on desktops, well, there's MythTV. But, if the developers can't figure out CableCard without violating the DMCA, then it won't happen. Some developers might figure something out, but it will never be part of the main codebase. Heck, they won't even allow talk about Torrentocracy on the myth-users mailing list because of the fear that MythTV will be seen as a bad program for allowing you to download TV shows.
All in all, it's a sad state of affairs, but unfortunately, that's the way it's going. At least in Europe, they have some idea on open standards, with things like DVB.
Myself though, I agree with you. I'm sticking with SDTV because I don't feel like paying a lot of money for a TV that will either try to lock me out of the content I'm paying for, or will be made obsolete with new DRM restrictions. I'm staying away from Digital Cable for the same reasons. But, unfortunately, most people just don't care.
-- Joe
My, my, we wouldn't be a disgruntled Dreamcast fan who waited for Half-Life, would we? :) (Or am I being trolled? I'm not really sure!)
Seriously though, them using Windows CE for Dreamcast played a major role. They couldn't keep a decent frame rate (which probably came from the combination of C++/WinCE/DirectX), and they were having trouble with sound. (The sound one I recall in particular because they wanted to do multiple effects on a playing sound, and DirectSound wasn't capable of it, because it was running an older sound driver that didn't support it. I think that the Sega sound driver supported it though, I just don't remember that far back).
Then there's the usual politics, return on investment questions, and the lead developer was apparently having personal problems which led him to taking some time off from work.
I guess that everything combined just caused them to drop the title. It clearly wasn't because of the Dreamcast hardware - the port of Quake III was quite good, and as I recall, it was one programmer who did everything.
-- Joe
I started at Sega just before the launch of the Dreamcast, so I don't know the entire story (or even if this one is completely true). Nonetheless, this is what I heard:
The team at SOJ (Sega of Japan) didn't want to make their own OS for the Dreamcast. They were uneasy about it, but chose Windows CE, as I guess they figured that it was similar enough to Win32. The fact that Microsoft was able to demo Internet Explorer on the Dreamcast was probably a big bonus, given Sega's desire to make the Dreamcast a big thing on the Internet.
However, as time went on, SOJ noticed that Windows CE was big and bloated and full of bugs. There was developer backlash. Sega's own software development teams (AM) needed something better, especially if they were to make full power of the Dreamcast (there was an arcade system that was basically a Dreamcast on steroids, although the name escapes me now). Thus, Sega of Japan started to develop their own low-level operating system for the Dreamcast. By the time that this happened though, the contract was in place with Microsoft for Windows CE - part of the contract was that Sega had to make WinCE available to all developers and stamp the logo on the unit.
My own observations:
Most of the developers in the United States and Europe used Sega's OS - it just provided the low-level functionality that tbey were used to. The developers who were using WinCE usually had an existing code base that ran on Win32 on the PC, and they were looking for a quick port. For games that weren't really intensive, this worked fine, but for some others (I saw early versions of Half-Life on the DC, when it was already delayed by at least six months, and let me tell you, it had *major* problems+) it was a disaster. What I would tell developers who were asking me if WinCE was worth using: "It will get you a solid 15 FPS, and if you're willing to optimize your rendering code with some assembly (to make use of the SH4's vector functions), you should be able to get 30 FPS. You'll get your game up and running faster, but you'll spend more time optimizing it for speed."
Version 1 of WinCE for Dreamcast was pretty buggy, version 2 fixed a lot of things, but even once they came out with the final version (I think 2.1), there were still lots of bugs. Developers were asking me if Microsoft would release WinCE 3, or at least fix some of the bugs. I tried to get the source for WinCE 2.1 for DC (so that I could try to at least maintain it myself for the developers) and got nowhere. Mind you, this was only a few months from Sega canning the Dreamcast anyway.
Microsoft did what they could to get developers to use WinCE on the Dreamcast. They'd send out promotional material to every new (and existing) Dreamcast developer which included a T-shirt, a Leatherman tool and of course, the WinCE SDK CD. We got a lot of thanks for the free tool and a T-shirt, while the CDs went into the garbage.
-- Joe
+ That's not to say that the sole decision to cancel Half-Life for the Dreamcast was because of WinCE performance. There were also some other issues outside of that, which I shall not discuss, but WinCE was a big factor.
Sorry, brain fart there. Focusing too hard on SQL, and not enough on posting.
I meant 2002 (and of course, their companion program for the Nomad Jukebox line, Notmad Explorer).
-- Joe
No, I mean Windows Autorun, which determines when a CD has been inserted.
My guess is that (and I'm not familiar with Windows APIs, so I don't know if this is even possible) iTunes registers an event handler for when a new CD is inserted, and that triggers it to do ripping.
I disable Windows Autorun because I happen to have a CD that won't let you listen to it's content without installing a stupid program (or so I think - Exact Audio Copy will see the tracks, but Windows Explorer insists on opening only the data portion, and WinAmp can't open the tracks either), and I'd rather not install that program. (BTW, no, I did not buy this CD, it was a gift from a relative of mine who isn't too technically inclined. And even worse, they don't say that the CD is copy protected, they just say that it's "web-enabled", or something like that. But, there is a lack of the CD Audio logos in the packaging/jewel case). Also, Autorun just annoys me.
The help in iTunes (or, at least a dialog box that popped up) says that with Autorun disabled, I can signal to it that a new CD has been put in - I need to click outside of the iTunes window, and then click in it again. But that doesn't seem to work. Again, not an issue, since I'm not using iTunes anyway (except to sync up video and pictures the odd time).
-- Joe
Try Anapod Explorer from Red Chair Software instead of iTunes (http://www.redchairsoftware.com/anapod/). I find it works quite well on my shiny new iPod Video (for music only, but I'm sure that they'll have video support in due time).
iTunes has huge problems, but the one that bugs me the most is that I can't seem to get it to rip CDs with Autorun disabled. But, now that I'm using Anapod Explorer, it's a non-issue, since I'm re-ripping my CD collection into FLAC, and Anapod will convert it to WAV or MP3 on the fly before uploading it to the iPod.
It's a quality program, and I've been using it since 1992 on my Creative Nomad Jukeboxes.
-- Joe
In Canada, there's certainly no shortage of tech workers, but there is a shortage of tech jobs (though, I hear that is improving now with a bunch of tech companies in and around the Waterloo, Ontario area).
The Canadian government is afraid of "brain drain", and rightfully so - a lot of their bright technical people are leaving for the U.S., because they can't find jobs in Canada in their chosen field. I left Canada six years ago to go to California, where I'm working as an embedded systems programmer. My first employer came to me - this after some six months of job hunting within Canada, before and after graduation. I had never given thought to leaving Canada until I got that offer, and it was certainly better than working in a computer store, or working phone lines.
In fact, two of my friends from high school also came down to California (although both are back in Canada now), and I've met lots of Canadians down here. They all say the same thing - no (non-minimum wage) jobs in Canada for them, so they came down here.
-- Joe
I live in Northern California, but it wouldn't surprise me if we have the same provider. My services are provided by CCI (the full name eludes me right now), which is based out of Riverside, CA.
I've been using it for a couple of weeks now, and it's not so bad.
-- Joe
I'm actually in the same situation, I moved into a new apartment complex where they do "repackaged DirecTV" also. It's actually not that bad, but the biggest problem that I have is that I never know whether the glitches in the video are because of the signal, or my PVR-350/PVR-500. That and Zap2It (used by my MythTV machine) has incorrect listings for some of the channels (because they're on an EST feed, but Zap2It thinks that they're PST channels).
The only problems that you'll really have are when the dish can't pick up the signal, and possibly when the content is shown (like I said, most of my channels are on an Eastern feed, but I'm in California). But, I just record the shows that I want to watch, and it works out well enough. And, as a bonus, I get my cable TV, phone service, and internet for cheaper than Comcast cable and Internet services.
-- Joe
Dammit, my HTML skills are lacking tonight...
The underperforming people would definitely try to get into the union (I'm sorry that I didn't explictly state that). They would try to unionize because then they collectively would be able to force the employer not to get rid of them. The above-average performers would be dead-set against the union.
Why would the lower-end quit the union? That union is the only way they'd have their jobs! While the above-average people have a proven track records, thus making them attractive to other employers, the below-average do not.
So, let's say that a tech union manages to materialize at a job. Here's what I predict would happen:
- The above-average performers would quit the union (which likely would mean changing jobs), because:
1. They'd be sick of their underperforming co-workers floating off of them.
2. Unions actually sometimes set maximum caps for productivity, in order to make their underachievers not look so bad.
- The lower performance employees would stay in the union, because the union guarantees their jobs.
- Employers cannot refuse employees right to unionize... So they'd probably close down shop, and reopen. But, they'd know who was part of the union, and refuse to hire those people (not because they were union, but because they were underachievers).
- Meanwhile, the more productive employees who quit the union have moved on to another job where the employers prune the deadwood, so that there's no incentive to unionize.
There are several good companies out there, and I work for one. We're a small company, but all of us know what we have to do, and we get it done. We've never even thought of unionizing because we're treated very well, and get paid very good wages.
After all, that is theoretically what unions are supposed to offer employers, higher quality employees.
Most unions do not do that. Most unions are leeches, taking their members' dues and not helping out the members. They also cripple business. And most of the better tech people know this and stay well clear of them.
-- Joe
I would certainly hope not (although I wouldn't like to be the person to test that), mainly because if our employers don't owe us two weeks notice, then why should we?
That said, I have worked in companies where they have always provided advance notice of layoffs, and I have always given two weeks notice, if nothing else, to not burn bridges (and in some cases, even helped them out after those two weeks with my new employers' permission).
-- Joe
Do you honestly think that underperforming employees who become part of a union would agree to that? No, in fact, I'd wager that many of the employees who would benefit from a tech union would be the poorly performing employees.
In fact, I'd even say that the reason we don't have tech unions are because of the better performing employees who refuse to unionize for just that reason.
Also, the rules aren't necessarily put in place by the employees, they're put in place by whoever controls the union. This would not likely be tech workers, but instead another group of people (i.e. the local union), which wouldn't know the first thing about what the "lower rung" should be.
-- Joe
I agree wholeheartedly with everything that you said. I specifically chose not to mention patents though, because there are things that you could learn, that while not necessarily patentable, do seperate your company from others. (Perhaps an efficient algorithm for filtering packets, which although is blatently obvious, your competitors haven't figured it out, or felt the need to figure it out due to lack of competetion.)
I disagree with the, "All you know, and all you create while you work with us belong to us" contracts, and anytime I see language like that, it puts up a red flag. I immediately try to negotiate that. (In fact, my previous employer had something like that in the contract, I explained that California State Law doesn't allow for that, and I wanted it removed. I also got them to put in a clause explicitly stating that anything I did in my own time belonged to me, just so that there was no confusion.)
I just have a policy of not leaving it up to the courts. If something in a contract is unclear, it needs to be made clear.
As to your point about rewards for big ideas, well, I can't say that I've been in that situation yet, so I can't really comment either way.
Like I said in another post, I'm lucky - I've only had one employer who has treated me poorly in my professional career (I've had some who treated me poorly before however).
-- Joe
I would never participate in a tech union, because given my experience over the last six years (four different companies, but smaller ones), I know that I could do better.
Unions served a very useful role when they came out, but I've talked to (former) unionized workers recently, and their common conclusion is that unions protect the weakest people best - those who can't fulfill their job responsilibities very well (i.e. meet the quota, and a reasonable one at that).
I'd rather perserve my job by showing management that I can work better than those people. It's worked for me very well thus far in my career, as I've increasingly been given more important roles and pay raises, while the lesser skilled people were shown the door.
An example of the crap I've seen: One software developer who spent three weeks on an issue, and every time when asked why it wasn't fixed, would say, "Hardware problem. I can't fix it". The boss would re-assign the bug to me, and I'd have it fixed in two days (without any of their bugtracking notes, since they never logged any in the system, so I came in cold) - definitely not a hardware problem, but an unskilled person.
Then again, as I've said, I've worked at smaller companies (at most 20 technical people in my group), and my employers were good enough to give me a chance to show my worth (and reward me for it). I'll admit that I've been very lucky in that regard. I also don't put in any more than 45 hours in a given week (unless an immediate problem comes up), and my manager is perfectly fine with that.
-- Joe
As rediculous as that sounds, that's not a bad idea.
If you can't understand the terms of any contract, you should definitely consult a lawyer before signing it. Also, I'd be wary of working for such a company anyway - a company that has to spell things out in such a manner (beit explicit or obfuscated) likely wouldn't be a company I'd want to work for anyway. I've found that I work better when I have fewer restrictions like that.
On the other hand, I dispise the fact that things have gotten so bad in employment.
-- Joe
No, this is why people should be very diligent in reading the fine print on contracts.
If you can't abide by the terms of the contract, then you either negotiate the contract, or don't take the job, it's that simple. If people wouldn't accept the jobs, then companies like Microsoft would have to treat their employees better.
Sidenote 1: From the time I spent at the Microsoft campus in Richmond (one week in late December 2000), I got the impression that full-time employees are treated quite well. Everybody has their own office (most people suffer in cubeville), and they are encouraged to make their office very personal - those who are musically inclined actually have their instruments there, for instance (well, smaller instruments). Of course, this is part of then encouraging them to work longer hours, but it wasn't forced to the best of my knowledge. Just avoid the cafeteria food, I got pretty sick off of some of it.
Sidenote 2: I wouldn't work for Microsoft, not that I couldn't have at least gotten an interview. I wouldn't work for them, just because many of the employees, while great people, are just corporate zombies - you flick a switch, and they say, "I can't understand why the government is out to get us, we've done nothing wrong!"
But that's totally unrelated to the point of my comment.
-- Joe
Did you mean non-compete or non-disclosure?
If you meant non-compete, then I agree with what you said. If you did mean non-disclosure, then I disagree.
In the case of a non-compete, since the company would be preventing me from working for another company in the field, I definitely should get a golden parachute. Should I obtain a position at a company that is not considered a competitor, the terms of the contract can be renegotiated.
Non-disclosures are a different beast. I feel that you shouldn't be able to use any proprietary information that you obtain at one company at another (for instance, you develop an algorithm that completely revolutionizes search engines, you can't give that to a new employer). Generic knowledge on the other hand, is fine. Even in search engine technology, there's a lot of general knowledge.
I live in California, and the NDAs that I have signed basically state that I won't take any company secrets with me to a new employer. General knowledge isn't considered company secrets by any extent, so I'm quite free to move (and move I have, I've changed companies four times in the last six years).
Most importantly, I read the paperwork before I sign it, and if I disagree with it, I negotiate. At the previous company I worked at, they didn't want to give me any vacation time for the first six months, and also wanted to only give me 10 days (the norm is 15). I explained to them that I was taking over 10 days of vacation time in three months, and if they didn't like that, they could look for somebody else (though not in those words). I still got the job, because they felt I was the best candidate.
-- Joe
To some extent that is true, but not always. The sticky issue is that of device drivers. Device drivers hook into the kernel, and their licensing as such has never really been clear. The general consensus seems to be that as long as you mark your module as non-GPL in the declaration, which excludes your use of some GPL'd code, your module is okay.
Userspace applications don't suffer from these restrictions - glibc is LGPLd, as is uClibc for these reasons - you don't have to abide by the GPL in order to have a C runtime library.
Where I see this going (as I stated in another post) is that whether Linux or BSD is used will likely depend on the hardware designers (companies like Broadcom, who make the reference designs, not companies like Linksys, D-Link or Netgear, who just base their work on reference designs).
The sky is not falling, I agree. We'll continue see substandard products from el-cheapo manufacturers no matter what underlying OS is used. I am going to go out on a limb and say that if FreeBSD can be used in these low-power, slower CPU, small RAM/flash footprint devices, and it performs as well as Linux, then the designers will do so. It gives them that much more protection against violations of the GPL (accidental or not).
-- Joe
I posted something to this effect on the original CNet article
I work in embedded development myself (previously video game consoles, then DOCSIS cable modems, now video equipment), so I've seen the shift from expensive proprietary systems (like vxWorks) to free (as in money) systems like Linux.
The proprietary systems typically have high up-front costs, along with a per-unit royalty, which inflates the cost of the devices. Linux allows for cheaper devices (whether or not the savings are passed to the customer remains to be seen), at a cost (complying with the GPL). This can be somewhat mitigated by making modules that are not licensed under the GPL.
BSD entering the space will provide some good competition for Linux. Whether newer designs switch to BSD will depend on the chipmakers (like Broadcom), as they are the people who usually write the drivers. Most devices nowadays are just the reference design hardware tweaked a bit with the reference software. So, whatever OS is used for the reference designs is what will be the dominant OS in the embedded space.
Only time will tell, but if FreeBSD can pull this off, they'll definitely gain some traction.
-- Joe
They honestly all are crap.
Nowadays because the margins on networking hardware are so small, Netgear, D-Link, et al basically take a reference design and slap a different web interface on it. There's no differentiation in the actual software that controls the wireless portions. Updates to the software come when the chipset vendor releases updates to their customers. Then to top it off, the hardware is manufactured in Taiwan with the cheapest possible components. (I worked for a Taiwanese company making DOCSIS cable modems, that's the way that they all operate).
However, the 2.4GHz band is absolutely awful. In my apartment complex, there's a half-dozen networks around me (802.11b/g), to the point where I was having interference issues with the other networks. My solution was to invest in 802.11a equipment, which is working well, although quite expensive (well, not really expensive, but twice the cost of your average 802.11g system).
I'm actually a fan of the Netgear client for my Netgear PCMCIA card. It works quite well after a Windows re-install, but the software definitely doesn't like having drivers/software around for about a half-dozen other cards around (testing compatibility with other cards).
-- Joe