Spinning off Agilent was actually a good idea. Terrible name but still a good business decision.
The problem was HP was a bit too big and covered too much areas, and had too many competitors. They couldn't sell spectrum analyzers to places like Dell, IBM, Compaq, because they had a PC line. And who buys from the competitors.
They also couldn't sell computers to places like Techtronics, Rhode and Swartz, and other scientific instrument places. Of course I mangled all those names with my spelling.
As a customer of HP, I never bought their computers in the first place. They were always overpriced. But I bought heavily on the Spectrum Analyzers, Sig Gens, etc. The worst thing for me in the spin off was the name. I still call the equipment I get from Agilent as HP. Its just easier to say, and old habbits die hard.
I don't think HP will be able to reaquire Agilent. They don't have the cash on hand. The stock holders probably won't go for it either. And the feeling I get from the Sales Reps I deal with it seems that Agilent looks at the renaments of HP with some scorn, and that's probably throughout all the company. And Agilent still makes some damn good equipment. If HP keeps going downhill despite the CEO leaving, I could see Agilent aquiring HP just to get the name back.
Automation is also good for complex embedded devices that are designed to an open specification. Along the same lines its also good for testing servers written to the specifications.
What white box automation needs (white box: where the testers are not contaminated by the code, and cannot contaminate the code) is a way to inject events into the device under test (DUT) and a way to observe and record repeatable results.
For complex embedded devices that communicate with other things (example cellphones, or USB) this is fairly easy. It is also easy for servers. You send a message to the DUT and expect the correct response. If you don't get the expected response you either mark it as a failure or fix the automation.
But this requires a specification that is agreed upon by both teams, and the development of the automated test system, is independent of the DUT. And this doesn't require a stable system to test on.
You can't test applications this way. At least not easily. You have to keep rewriting the automation to 'click' in different places as the GUI changes.
Especially in a discussion talking about embedded systems. Though people who use flash seem to have no sense of the tightness needed for small bandwidth systems.
For example: (and I am an embedded programmer, who see functionality, speed, and algorithms as more important then flash) I help run a website for a set of international martial arts games. The webserver sits on an ADSL link, that charges $25/gig over the first gig upstream.
One of the (new) website developers said: I can make a nice fancy flash image for the first page. It'll be very small. the rest of us: how small? Him: Only about a meg.
So 4 megs might not be that heavy afterall. To me its heavy, to anyone who does embedded programming, its heavy. Its just a different mindset.
But a bit back on topic for the general article. If you look at things, WinCE and Linux are probably too burdensen for most embedded devices. Coming from trying to switch out 7 year old ARM development tools for the GCC 3.3 tool chain I've come to realize that alot of open source stuff isn't going to hit embedded devices until they get more people who are into the small, tight, fast mindset. I was taking a set of code that made a 1.5 meg flash image and trying to get it to compile/link with gnu. With all the size optimizations, strip commands (even more then just doing strip, but taking away whole sections like the.comment section) the final code turned out to be 3.9 megs. When you look at it, the difference in price for a million units of 2 meg flash chips and 4 meg flash chips, didn't justify not spending money on the Arm development system.
So you aren't going to see linux on the small devices until you can get the gnu tool chain to compile smaller code.
Verizon is probably just pissed because they don't like competition. As for the 1.6 vs 3 billion, I'm not sure its that big of a steal. The 800 MHz mobile band is much more valuable then the 1.8 GHz PCS band, from the providors point of view.
One of the rules of RF cost/design is that as frequency goes up the electronics to increase the power and quality of the signal goes up, while the cost/size of the antennia goes down. The problem is, that the cost goes up faster for the electronics then the cost goes down for the antennia.
So they gave up an 800 MHz block that would probably be worth at least 1.5 to 3 billion to them, and bought another block. The money almost works out the same. What would have really sucked for Nextel and probably what Verizon wanted was for them to give up the 800 MHz block. And then lose the auction for the new block, cutting Nextel out of the market and reducing competition.
Also the higher the frequency the worse it is for distance. I can't remember the exact figures but I think for PCS you need 3 - 4 times the number of base stations as you need for the 800 MHz band.
As for upgrading the customer's phones, it won't be much if they already support the new band as many newer phones do. If they don't the customers are going to have to get new phones. In BC when Telus upgraded their PCS base stations to 3G the old 2G phones wouldn't work. They gave out their cheep phones to the old customers and didn't make them get a new contract. Nextel can do the same thing.
For the phones that already support the new frequency. The new frequency plan can be downloaded over the air during off hours, or even when the customer is talking on the phone.
I would suggest for anyone who likes British Comedy and wants to see how some of how politics and backroom dealing works, look for 'Yes Minister' and 'Yes Prime Minister'.
The quote that I find most relevant here, And I'm going from memory is:
Prime Minister: I explicitly ordered the UN ambassidor to abstain from the vote!
Humphery: Yes but an order to the Foreign minster becomes a strongly worded word to the permanent secretary. Which becomes an advisory to the UN under secretary. Which becomes a suggestion to the Ambassador. Who will vote for what's best for him.
I would not be surprised if this thing happened in the countries involved. Yes Minister while being a satire and funny. Is still a satire and reflects on real life.
Don't forget that since they're 'giving away' the hardware they are going to wessel out of being responsible for it. And want to bet that if that hardware breaks then you'll have to pay for a new set, which won't be included in the software subscription.
Hey I'm an engineer, I'm allowed to be cynical. Ontop of that I work for a cash starved start-up that hasn't folded yet, just watching the 'marketing' department work in cases like that shows just how close to the truth a cynic is.
Don't know if you're serious or not. But this is a bad idea.
A lot of accidents (I don't know the percentage but I'm sure its most) happen at intercetions. And that usually means traffic lights. Ejection seats would have a very good chance of shooting a guy right up into one. Not to mention trees, buildings, and a whole bunch of other things that are along the side of the roadway.
Another problem is that if any part of the system fails then the ejector is a going to be in a world of hurt. One of the many reasons a combat pilot wears a crash helmet in his fighter is because if he has to eject and the canopy doesn't go, the helmet helps him get through the glass. You're not going to get everyone in cars wearing crash helmets.
Say the car is flipped over and the ejector is automatically turned off, and the car is onfire. You now have a person strapped into a chair with alot of very explosive rocket fuel right next to his butt.
That brings up the nice nifty seatbelts. The standard lap and shoulder restraints aren't going to do for an ejector seat, and we have enough problems getting people to use the seatbelts as they are.
Also while driving a person's seat is usually partially under the steering wheel, and the legs under the dashboard to get to the pedals. An auto ejection won't allow a person to clear those obsticals before he is rocketed through them.
Multiple people in the car will also be a problem. You can't fire the seats all at once because the parachutes would get fouled on each other. But then you can't fire them independently because the people remaining in the car will get burned by the rocket exhaust. You can't fire them all of in different angles because that increases the chance of lauching a person into a building, or high voltage line.
I'm sure there are more problems with putting ejector seats in cars then these.
CDMA is also a complete radio protocol. I have the misfortune of developing the link layer and control layer for it. This is the software that handles making calls and getting the CDMA phone on the network.
The CDMA radio interface or physical layer is better then the TDMA layer that GSM uses. From some courses I took, it worked out that CDMA allows for about 5 times the capitity on the same bandwidth that GSM/TDMA allows, that's the practical limitation.
But the software layer for link and control layers is a MESS, a true and undisguised mess. Now I haven't seen GSM or W-CDMA's link and control layer, but nothing can be as bad as CDMAs link and control layers. This is because there are ten different protocol revisions from JSTD.0008 to IS-2000 Revision C. I don't know how bad GSM->W-CDMA is, but W-CDMA's physical layer is not backwards compatible with GSM so it doesn't have the problem of backwards compatibility that CDMA requires.
TD-CDMA is a completely different beast of course. I think is the same as TD-SCDMA that China has come up with, which seems to be a mash of CDMA, W-CDMA, SCDMA, and a few others. But it uses the W-CDMA messaging and control with some modifications.
Alot of the reasons that people use to claim that GSM is better then CDMA is based on it being an older technology, with 10 more years of development behind it. 10 years gives you smaller/cheeper chips which provides cheeper products. It also allows for ten more years of applications and add-on developments. And of course the APIs between CDMA and GSM are different so you can't port that across easily. And 10 years gives you a lot of market penitration.
Also look at how your swap file is set up. If its set to dynamic you might be screwed. As you run your system and you need more and more memory the swap file grows, and gets fragmented. The defragger that comes with W2K doesn't defrag it. You need to use something like diskeeper to defrag it, and it usually helps speed the computer back up.
This is the only obvious thing to me that makes the system go slower as time goes on, that doesn't get fixed by a reboot. There's probably hundreds of other things that cause it but I'm not sure what they are.
Its vapourware and IMO it will never be anything but. If you look at some of the financial articles on Hop-on they come to the conclusion that Hop-on is just a way to get money from investors. They haven't produced anything and probably never will.
They're president (Peter Michaels) had to step down because he was arrested by the SEC I believe. I don't remember exactly what happened but he was arrested because of some of the scams he was pulling.
They work with cellphone chip set manufacturers and then screw them just before going to production and canceling the project. They had a phone get through FCC for CDMA, and they dropped it, even though the software was done for the phone and it passed the compiance testing. This was over a year ago, and they haven't produced anything but press releases. Though no-one seems to be biting they're stock since it has been ~0.10 for a year unlike SCO.
And the slashdot story doesn't point to anywhere at all, but to the hop-on website as if it was new. Its been on slashdot so often it isn't funny anymore.
Exactly. There's nothing you can do about stopping someone from emailing a virus. You can stop it at an email gateway of course, but nothings 100%. I accept that.
What I don't accept is virus that are automatically executed simply by viewing an email in the preview pane. As soon as you click on it, you're infected.
We've mostly got visual basic scripting to thank for that.
I have to agree. What really gets me is I see the title of the message, right click on it so I can delete it, before I can delete it the virus scanner on my machine goes off and the menu goes away. A mail goes to my IT support and they yell at me for downloading viruses. And I get something like 20-40 of these a day.
And the rules in outlook to delete the messages don't work worth anything. Most of them say they're from microsoft. So I set up a rule to say 'if its from microsoft and it contains an attachment delete it' but does this work? No. Also alot of them say 'here is the qmail program' and I have a rule to turf those, but it only gets about half of them.
There is nothing you can do about these viruses as a user of an exchange server with Outlook. But we have to use it for meetings and resource scheduling, which is a piece of crap! Microsoft has almost a monopoly on this in small to meduim bussinesses. I've also used two of the other big time mail/scheduling software (lotus notes and novell groupwise) and they're crap too. But we can't use gnu in the office right now, damn SCO.
Ummm...CDMA2000 is more of a world standard right now than WCDMA is. You can pick up a CDMA 2000 signal in all of the US, most of Asia, and just about all of South America! Right now, you can only find WCDMA signals in a few areas of Europe, and Japan.
IMHO, by legislating what cell phone standards are being used, Europeans are possibly allowing buerocrats to decide what technology is better!
[Marketing Talk] But WCDMA is going to be big Big BIG!!! [/Marketing Talk]
I work in the cellphone industry and I don't know who's bigger right now and who's going to be bigger in the long run. When I first started out I thought it was going to be CDMA-2000. Then I saw some of the monopolistic and protectionist tactics of the company with most of the tatics force away most of the people wanting to do CDMA-2000. TI said they were going to start making a CDMA-2000 chip and got slapped with a lawsuit within days. I think a few others have as well. OF course these are just what I hear from our marketing people who are useless.
Then I thought it would be W-CDMA would be the way to go, because it uses the GSM upper layers, which I hear are easier to implement and you can get SDL models for. I know for a fact that W-CDMA's over the air messages (upper layer), are a lot cleaner then CDMA-2000, even though I haven't looked at them. Nothing, I mean NOTHING can be as convoluted and horrible to encode and decode as CDMA-2000's over the air messaging (if you don't beleive me have a look at the C.S0005-C document on www.3gpp2.org and look at the Extended Channel Assignment message, and the General Handoff Message). But the same company that has patents on most of CDMA-2000's stuff has some key patents on W-CDMA. And W-CDMA has some technical issues in the physical layer that have mostly been worked out.
Now there's of course the TD-SCDMA that's coming out of China. There is different company in China that holds the patents, and they're part of a gov't consortium that allows others in that consortium access. I don't know the fees or what not. But the consortium members must have a chinese presence (I'm pretty sure they have to be Chinese owned, but I'm not sure), and TD-SCDMA research is funded by the Chinese gov't and they are trying to push it over the other standards. That's alot of market pressure, and I don't think the Chinese gov't would allow its companies to do protectionist, and monopolisitc practices as the american company does to other chinese companies in the consortium. Other companies are fair ball of course.
In the future you might find that TD-SCDMA will come to the front because the technology is cheeper. I have no idea if it will be better. But it will come from China where labour and research is cheep compared to the rest of the world.
$30,000 - $50,000 is about what it's going to cost to employ another full-time programmer in a developed country.
Salary counts for less than half of the total costs of having a salaried employee. There are taxes, benefits, building costs, equipment costs, training, etc.
Do you really think that they can get good technical employees who are willing to work for US minimum wage (and probably less than minimum wage in France)?
To give a good example of this. Back when I was well paid, and I use to deal with engineering burden rates for figuring out the cost of my time in dollars for upper management, I had a set figure, that was $625 US a man day. This was for the canadian office.
If you took the average salary at the canadian office you'd see it work out to something like this:
Engineering Cost per year $162,500 Engineering Salary per year 45,360 All the rest: 117,410 (72%)
A dozen engineers would cost: $1,950,000
By the grand parents numbers of advertising (which I don't have on my screen so I can't see them exactly Damn I think its 30k per year per advertizer) Mandrake needs 65 advertisers just to handle the 12 engineers.
Of course part of the burden (appart from the engineer's salary) is the salary for 'non-production' personal, or non-billing personal. This is all your C*Os, secretaries, and marketing. It does not include production costs, those are usually covered under a different system.
But welcome to the world of 'too much information, more then I wanted to know' world of project management and budgeting.
Re:Let's make this a press release!
on
Back To SCO
·
· Score: 1
Why is it that there isn't a push one way or the other to amend the constitution to clarify the "right to bear arms"?
Being a Canadian I always thought this was a better interpretation. You have all the right to go out into the woods with a pen knife and get all the bear arms you like.
Or just interpretate it to mean you're allowed to wear short sleeves. But this would make all the $5 T-shirt producers, who put something on them and sell them for $25, happy.
So, hollywierd is upset that word of mouth is traveling faster than ever. Why don't they start making movies that don't suck? As for EULAs. I didn't sign anything. I didn't agree to anything. They can put them up there if they want to.
Then they'll start forcing you to pay by credit card. And on the little bill right by the signature portion they'll put in super fine print... "By signing this you agree to abide by the EULA"
Of course for them to do this they'll have to find a way to get people to stop paying cash. All they have to do is up the price a bit more. Its already getting to the point where I don't go to the movies because its too expensive and when I do go I'm hesitant giving over cash because I don't carry that much with me anymore.
They could also weasle it in as 'Oh won't someone think of the children' by making R rated movies require the credit card as proof you're an adult. A driver's license won't do because movie theatres will probably get rid of cashiers to look at them, and they won't be hooked into the licensing network for those of us who have magnetic strips on their DLs.
This collapse was predicted years and years ago. When I was working for a power equipment manufacturer (transformers, relays, switches, capacitors), marketing was playing this sort of thing up to the utilities.
As equipment gets old it becomes less and less efficient. This includes the transformers that bring the power from high voltages to low voltages to your home, and the generators that produce the power. AND it includes all those Air conditioners that are running in hot weather.
No-one has been able to afford to bring new generators online recently. And probably not to upgrade/replace old less efficient equipment. And I'm sure most people haven't bought new ACs either because of the economy.
It wasn't so long ago that something similar happened to Chicago during a heat wave there. And we all remember a few years ago that California had rolling blackouts because the grid couldn't handle the power. And NYC suffered similar blackouts for the same reason in the 60s and 70s I believe.
Unfortunately since its such a large grid its going to take a while for it to come back up. You have lots and lots of main power generators. Each one has to be brought back onto the grid one at a time. Each one has to be synced to the current AC 3 phase system within 5-10 degrees of what's there or when that generator comes on it might cause all the generators to drop out. Syncing a generator takes time and patience.
Then you have to bring the consumers back on. Every time you bring a new section on you have a hell of alot of inrush current as Air Conditioners and motors start up. This is why your lights dim a bit when you turn on certain pieces of equipment. Imagine the dimming you get as 1000 Air conditioners come on at once. If its too much a relay might trip off and the grid might collapse under the strain as a generator falls offline. And yes this is a real meaning to the word offline, the generator is not on the power lines anymore.
It will take time for everything to come back up, and con-ed isn't going to rush it. They will take it up in stages, make sure that everything is ready to go before bringing up the next stage. A collapse this large can happen again and again if they rush. But it might be quicker, the reports don't saay how big the failure was and how many generators fell offline. It could just be that all the distribution substations tripped, but I doubt it. For this to be so widespread the generator protection relays probably all fired off and took their generators off the power grid.
Don't you just love cascading failures? Overloaded power grid; all the generators are close to their shut off point. One fails, all generators go into the range of shutoff, and off they go one right after the other. They probably all fell offline withing 30 seconds, and will probably take 3 days to come back on fully.
Sounds like your power supply can't handle the inrush current for the start up of your hard drives and fans.
When motors start up they have to get over the standing still friction of the unit before they get over the hump of startup and then into the lower friction of a moving unit. I would put in a nice graph but slashdot won't let me.
What the graph looks like is there's a rise up to a peak. Doing an integration under the line to the peak gives you the force needed to get to the peak. This is the force needed to get to the object moving. The graph then falls down a fair ways, (maybe almost to zero it depends on the two materials being being rubbed together), and then the line stays steady, horizontally. This represents the force needed to keep the object moving.
This is about all I remember from College physics on friction. There's also some other things that require a high inrush current during startup. But motors are the biggest thing.
So when you rock your power switch just right you leave the motors running enough not to require the inrush current. Shut off power to the motherboard to force it to restart, but you don't leave the power off long enough to cause the power supply to go into suspend mode and turn off all the power to the system.
Probably the best thing to do is get a power supply that handles a larger inrush current then the current one. But there could be something else wrong with the system too, and I haven't built a system for close to 8 years and my knowledge might be a bit dated.
According to the latest GamePro, the "CD-only version of Sony's own system [was] internally named 'Play Station X' (which is why you sometimes see the abbreviation 'PSX' for original PlayStation stuff)." So basically, that's all there is to the PSX name. As far as PS1 and PS2, I'd imagine it's just so you can tell the original playstation apart from the revamped one and from the PS2.
You won't believe how often X makes it into marketing names because sometimes marketing people have no idea on names. You'd think they would, but they don't.
I use to work for a company that named their products Form [number] or something like that. With the initials F1, F2 and so on through the generations. It wasn't a very imaginative name to begin with. There was the F3 on the market and the engineers started work on the new product to be called the F4. But before that product hit the market or even started marketing it there was a need for an upgraded F3. So the engineers had no idea what to call it, lots of humming and hawing went around the place. The drafters got fed up with not knowing what to title their diagrams and put it into the system as FX, fully intending to go back and fix it later when the marketing department came up with a name. They different, and that company still makes the FX, even though they make the F4, F5 and now the F6.
The PSX name probably came out for the same reason. Sometimes Marketers can't come up with names, and give up too easily and just use the internal names. Which are usually confusing and unsuitable for the mass market. I'm not saying coming up with a suitable name is easy, but that's what the marketing department gets payed for.
I don't think there can be an appeal on an innocent verdict in a criminal trial in the states.
I think its called the double bind clause in their bill of rights. If you're charged for murder and get found innocent you can't be brought to trial again just because they found some new evidence.
I know in Canada that an innocent verdict can be appealed. And in non-criminal cases the double bind clause doesn't apply. But this was a criminal copyright trial wasn't it?
That's not as far off base as you'd think. Crientien (or how ever you spell his name) signed an agreement with Bush about mutual protection or whatver between US and Canada.
Basically it states that the US can invade Canada if there is an emergency. The agreement leaves out who declares the emergency (Bust probably), or what the emergency is (Canada is selling energy for a profit, probably), (Or there's a pirate in Inovac, and the only way to get there is to go through the rest of Canada).
As a Canadian, this just sickens me. But then again since when did anyone listen to the voters.
There is a hard limit on 2G CDMA of 61 cell phones per base station. But the caculations I was mentioning before (damn wish I still knew where they were), assumed a Frame Error rate of 2%. That still provides a reasonable voice quality rate. 1xRTT (the 3G that sprint and verison are using) has the same limitations on codes to use for cellphones. (That's not entirely correct, but I doubt that sprint and verison are using the more advanced 3G features that allow them to boost capacity without boosting bandwidth. I don't think most customers phones support these features yet)
There are also a whole bunch of other issues, dealing with what vocoder is being used, which also determine both voice quality and the ability to do error correction. The '8k' vocoders have better error correction then the '13k' vocoders but they sound like shit.
I don't know if Verison or Sprint are doing the same thing as Telus is doing. But Telus defaults to the 8k vocoders for all calls, except the calls to their support lines that are at 13k. Hmmm I wonder why? (cynism)
From the sounds of things both Sprint and Verison are running at maybe 5% FER and maybe 55-60 calls per base station. You'd have to look at the signalling traffic to be sure.
CDMA is also generally reguarded as more advanced outside of the US as well. Which is why the air access portion of CDMA is the basis for most 3G standards. Going forward GSM providors are moving towards UMTS (or is it another acronym starting with U, I don't remember). And that's basically another name for W-CDMA.
Its a bit of a mess of why there are divergent paths. But the big one, is that Qualcomm owns most of the patents on CDMA, and are really the only player for CDMA base band chips. I only know of one other company that has had their chips certified by Nortel and Lucient to work off their base stations, and they're just getting into production now.
GSM was out in the market first. Europe had alot of problems with 1G (analog) cell phones. There were so many different standards that you had the same problem moving form one system to another, that you have in the US. When 2G came along GSM was choosen for the most part to hold back these woes. I don't remember fully but I think GSM came out in the mid 80s. Qualcomm introduced CDMA in the early to mid 90s, about 10 years behind GSM. So GSM had a long time to get entrenched.
But GSM is alot like TDMA, at least at the physical layer. CDMA is completely different. Basically TDMA has the same limit as AMPS does. You can really only have one transmitter on one frequency at a given time. CDMA changes this, allowing multiple transmitters to be on all the time on the same frequency. Theroitically this means that CDMA is interferance limited to the number of cellphones talking to a base station, while TDMA/GSM is still frequency limited. I don't remember the exact numbers, but it works out to around 8 GSM phones can effectively share the same frequency, while 43-47 CDMA phones can share the same amount of frequency. Its a complicated formula, because CDMA uses more spectrum per signal, but you can have base stations sitting almost ontop of each other physically transmitting on the same frequency. These numbers are for 2G CDMA. 3G CDMA is suppose to increase capacity by 1.7 times.
Of course Europeans have a burr up their saddles when it comes to technology from the States (and the US has the same burr it seems). And Qualcomm isn't helping manners. Qualcomm is greeder then Microsoft, and are probably impedding the adoption of CDMA more then anything else because of their monopolistic attitude (Think threats to customers who consider using other chip sets, exhorbadant license fees, etc). True 3G CDMA does meet all the requirements of high speed and higher capactity. Sprint and Verison are not deploying true 3G CDMA. True 3G CDMA has a bandwidth an effective bandwidth of 5 MHz, just like W-CDMA has. Instead S&V are deploying 1xRTT which is backwards compatable with 2G cellphones. Don't ask me why CDMA-2000 and W-CDMA are not compatible with each other, its all infighting between the GSM manufacturers and Qualcomm. The basis of it is that if CDMA-2000 was the 3G standard, and backwards compatible with 2G CDMA, then why would new carriers want to buy GSM equipment if they're going to have to get rid of them later, why not just buy 2G CDMA and upgrade to 3G later.
But this is the cellular market place. If they can get more money out of their customers for less service they will. And this sort of attitude isn't at the provider/network level but basically goes all the way up to the base station manufacturers who run the international standards committees.
Slashdot Mirror
Spinning off Agilent was actually a good idea. Terrible name but still a good business decision.
The problem was HP was a bit too big and covered too much areas, and had too many competitors. They couldn't sell spectrum analyzers to places like Dell, IBM, Compaq, because they had a PC line. And who buys from the competitors.
They also couldn't sell computers to places like Techtronics, Rhode and Swartz, and other scientific instrument places. Of course I mangled all those names with my spelling.
As a customer of HP, I never bought their computers in the first place. They were always overpriced. But I bought heavily on the Spectrum Analyzers, Sig Gens, etc. The worst thing for me in the spin off was the name. I still call the equipment I get from Agilent as HP. Its just easier to say, and old habbits die hard.
I don't think HP will be able to reaquire Agilent. They don't have the cash on hand. The stock holders probably won't go for it either. And the feeling I get from the Sales Reps I deal with it seems that Agilent looks at the renaments of HP with some scorn, and that's probably throughout all the company. And Agilent still makes some damn good equipment. If HP keeps going downhill despite the CEO leaving, I could see Agilent aquiring HP just to get the name back.
Nautical Mile: 6000 ft.
Hour: 60 minutes
1 knot speed = 100 ft per minute.
There really isn't anything suitable in metric that doesn't break the system, unless you try metric time, and that hasn't worked in the past.
You could try to define:
Nautical km: 3600 m
Which makes the conversion to m/s easy. But as I said that breaks the whole fundamentals of the metric system.
Automation is also good for complex embedded devices that are designed to an open specification. Along the same lines its also good for testing servers written to the specifications.
What white box automation needs (white box: where the testers are not contaminated by the code, and cannot contaminate the code) is a way to inject events into the device under test (DUT) and a way to observe and record repeatable results.
For complex embedded devices that communicate with other things (example cellphones, or USB) this is fairly easy. It is also easy for servers. You send a message to the DUT and expect the correct response. If you don't get the expected response you either mark it as a failure or fix the automation.
But this requires a specification that is agreed upon by both teams, and the development of the automated test system, is independent of the DUT. And this doesn't require a stable system to test on.
You can't test applications this way. At least not easily. You have to keep rewriting the automation to 'click' in different places as the GUI changes.
Especially in a discussion talking about embedded systems. Though people who use flash seem to have no sense of the tightness needed for small bandwidth systems.
.comment section) the final code turned out to be 3.9 megs. When you look at it, the difference in price for a million units of 2 meg flash chips and 4 meg flash chips, didn't justify not spending money on the Arm development system.
For example: (and I am an embedded programmer, who see functionality, speed, and algorithms as more important then flash) I help run a website for a set of international martial arts games. The webserver sits on an ADSL link, that charges $25/gig over the first gig upstream.
One of the (new) website developers said: I can make a nice fancy flash image for the first page. It'll be very small.
the rest of us: how small?
Him: Only about a meg.
So 4 megs might not be that heavy afterall. To me its heavy, to anyone who does embedded programming, its heavy. Its just a different mindset.
But a bit back on topic for the general article. If you look at things, WinCE and Linux are probably too burdensen for most embedded devices. Coming from trying to switch out 7 year old ARM development tools for the GCC 3.3 tool chain I've come to realize that alot of open source stuff isn't going to hit embedded devices until they get more people who are into the small, tight, fast mindset. I was taking a set of code that made a 1.5 meg flash image and trying to get it to compile/link with gnu. With all the size optimizations, strip commands (even more then just doing strip, but taking away whole sections like the
So you aren't going to see linux on the small devices until you can get the gnu tool chain to compile smaller code.
Verizon is probably just pissed because they don't like competition. As for the 1.6 vs 3 billion, I'm not sure its that big of a steal. The 800 MHz mobile band is much more valuable then the 1.8 GHz PCS band, from the providors point of view.
One of the rules of RF cost/design is that as frequency goes up the electronics to increase the power and quality of the signal goes up, while the cost/size of the antennia goes down. The problem is, that the cost goes up faster for the electronics then the cost goes down for the antennia.
So they gave up an 800 MHz block that would probably be worth at least 1.5 to 3 billion to them, and bought another block. The money almost works out the same. What would have really sucked for Nextel and probably what Verizon wanted was for them to give up the 800 MHz block. And then lose the auction for the new block, cutting Nextel out of the market and reducing competition.
Also the higher the frequency the worse it is for distance. I can't remember the exact figures but I think for PCS you need 3 - 4 times the number of base stations as you need for the 800 MHz band.
As for upgrading the customer's phones, it won't be much if they already support the new band as many newer phones do. If they don't the customers are going to have to get new phones. In BC when Telus upgraded their PCS base stations to 3G the old 2G phones wouldn't work. They gave out their cheep phones to the old customers and didn't make them get a new contract. Nextel can do the same thing.
For the phones that already support the new frequency. The new frequency plan can be downloaded over the air during off hours, or even when the customer is talking on the phone.
I would suggest for anyone who likes British Comedy and wants to see how some of how politics and backroom dealing works, look for 'Yes Minister' and 'Yes Prime Minister'.
The quote that I find most relevant here, And I'm going from memory is:
Prime Minister: I explicitly ordered the UN ambassidor to abstain from the vote!
Humphery: Yes but an order to the Foreign minster becomes a strongly worded word to the permanent secretary. Which becomes an advisory to the UN under secretary. Which becomes a suggestion to the Ambassador. Who will vote for what's best for him.
I would not be surprised if this thing happened in the countries involved. Yes Minister while being a satire and funny. Is still a satire and reflects on real life.
Don't forget that since they're 'giving away' the hardware they are going to wessel out of being responsible for it. And want to bet that if that hardware breaks then you'll have to pay for a new set, which won't be included in the software subscription.
Hey I'm an engineer, I'm allowed to be cynical. Ontop of that I work for a cash starved start-up that hasn't folded yet, just watching the 'marketing' department work in cases like that shows just how close to the truth a cynic is.
Don't know if you're serious or not. But this is a bad idea.
A lot of accidents (I don't know the percentage but I'm sure its most) happen at intercetions. And that usually means traffic lights. Ejection seats would have a very good chance of shooting a guy right up into one. Not to mention trees, buildings, and a whole bunch of other things that are along the side of the roadway.
Another problem is that if any part of the system fails then the ejector is a going to be in a world of hurt. One of the many reasons a combat pilot wears a crash helmet in his fighter is because if he has to eject and the canopy doesn't go, the helmet helps him get through the glass. You're not going to get everyone in cars wearing crash helmets.
Say the car is flipped over and the ejector is automatically turned off, and the car is onfire. You now have a person strapped into a chair with alot of very explosive rocket fuel right next to his butt.
That brings up the nice nifty seatbelts. The standard lap and shoulder restraints aren't going to do for an ejector seat, and we have enough problems getting people to use the seatbelts as they are.
Also while driving a person's seat is usually partially under the steering wheel, and the legs under the dashboard to get to the pedals. An auto ejection won't allow a person to clear those obsticals before he is rocketed through them.
Multiple people in the car will also be a problem. You can't fire the seats all at once because the parachutes would get fouled on each other. But then you can't fire them independently because the people remaining in the car will get burned by the rocket exhaust. You can't fire them all of in different angles because that increases the chance of lauching a person into a building, or high voltage line.
I'm sure there are more problems with putting ejector seats in cars then these.
CDMA is also a complete radio protocol. I have the misfortune of developing the link layer and control layer for it. This is the software that handles making calls and getting the CDMA phone on the network.
The CDMA radio interface or physical layer is better then the TDMA layer that GSM uses. From some courses I took, it worked out that CDMA allows for about 5 times the capitity on the same bandwidth that GSM/TDMA allows, that's the practical limitation.
But the software layer for link and control layers is a MESS, a true and undisguised mess. Now I haven't seen GSM or W-CDMA's link and control layer, but nothing can be as bad as CDMAs link and control layers. This is because there are ten different protocol revisions from JSTD.0008 to IS-2000 Revision C. I don't know how bad GSM->W-CDMA is, but W-CDMA's physical layer is not backwards compatible with GSM so it doesn't have the problem of backwards compatibility that CDMA requires.
TD-CDMA is a completely different beast of course. I think is the same as TD-SCDMA that China has come up with, which seems to be a mash of CDMA, W-CDMA, SCDMA, and a few others. But it uses the W-CDMA messaging and control with some modifications.
Alot of the reasons that people use to claim that GSM is better then CDMA is based on it being an older technology, with 10 more years of development behind it. 10 years gives you smaller/cheeper chips which provides cheeper products. It also allows for ten more years of applications and add-on developments. And of course the APIs between CDMA and GSM are different so you can't port that across easily. And 10 years gives you a lot of market penitration.
Also look at how your swap file is set up. If its set to dynamic you might be screwed. As you run your system and you need more and more memory the swap file grows, and gets fragmented. The defragger that comes with W2K doesn't defrag it. You need to use something like diskeeper to defrag it, and it usually helps speed the computer back up.
This is the only obvious thing to me that makes the system go slower as time goes on, that doesn't get fixed by a reboot. There's probably hundreds of other things that cause it but I'm not sure what they are.
Its vapourware and IMO it will never be anything but. If you look at some of the financial articles on Hop-on they come to the conclusion that Hop-on is just a way to get money from investors. They haven't produced anything and probably never will.
They're president (Peter Michaels) had to step down because he was arrested by the SEC I believe. I don't remember exactly what happened but he was arrested because of some of the scams he was pulling.
They work with cellphone chip set manufacturers and then screw them just before going to production and canceling the project. They had a phone get through FCC for CDMA, and they dropped it, even though the software was done for the phone and it passed the compiance testing. This was over a year ago, and they haven't produced anything but press releases. Though no-one seems to be biting they're stock since it has been ~0.10 for a year unlike SCO.
And the slashdot story doesn't point to anywhere at all, but to the hop-on website as if it was new. Its been on slashdot so often it isn't funny anymore.
Exactly. There's nothing you can do about stopping someone from emailing a virus. You can stop it at an email gateway of course, but nothings 100%. I accept that.
What I don't accept is virus that are automatically executed simply by viewing an email in the preview pane. As soon as you click on it, you're infected.
We've mostly got visual basic scripting to thank for that.
I have to agree. What really gets me is I see the title of the message, right click on it so I can delete it, before I can delete it the virus scanner on my machine goes off and the menu goes away. A mail goes to my IT support and they yell at me for downloading viruses. And I get something like 20-40 of these a day.
And the rules in outlook to delete the messages don't work worth anything. Most of them say they're from microsoft. So I set up a rule to say 'if its from microsoft and it contains an attachment delete it' but does this work? No. Also alot of them say 'here is the qmail program' and I have a rule to turf those, but it only gets about half of them.
There is nothing you can do about these viruses as a user of an exchange server with Outlook. But we have to use it for meetings and resource scheduling, which is a piece of crap! Microsoft has almost a monopoly on this in small to meduim bussinesses. I've also used two of the other big time mail/scheduling software (lotus notes and novell groupwise) and they're crap too. But we can't use gnu in the office right now, damn SCO.
Ummm...CDMA2000 is more of a world standard right now than WCDMA is. You can pick up a CDMA 2000 signal in all of the US, most of Asia, and just about all of South America! Right now, you can only find WCDMA signals in a few areas of Europe, and Japan.
IMHO, by legislating what cell phone standards are being used, Europeans are possibly allowing buerocrats to decide what technology is better!
[Marketing Talk] But WCDMA is going to be big Big BIG!!! [/Marketing Talk]
I work in the cellphone industry and I don't know who's bigger right now and who's going to be bigger in the long run. When I first started out I thought it was going to be CDMA-2000. Then I saw some of the monopolistic and protectionist tactics of the company with most of the tatics force away most of the people wanting to do CDMA-2000. TI said they were going to start making a CDMA-2000 chip and got slapped with a lawsuit within days. I think a few others have as well. OF course these are just what I hear from our marketing people who are useless.
Then I thought it would be W-CDMA would be the way to go, because it uses the GSM upper layers, which I hear are easier to implement and you can get SDL models for. I know for a fact that W-CDMA's over the air messages (upper layer), are a lot cleaner then CDMA-2000, even though I haven't looked at them. Nothing, I mean NOTHING can be as convoluted and horrible to encode and decode as CDMA-2000's over the air messaging (if you don't beleive me have a look at the C.S0005-C document on www.3gpp2.org and look at the Extended Channel Assignment message, and the General Handoff Message). But the same company that has patents on most of CDMA-2000's stuff has some key patents on W-CDMA. And W-CDMA has some technical issues in the physical layer that have mostly been worked out.
Now there's of course the TD-SCDMA that's coming out of China. There is different company in China that holds the patents, and they're part of a gov't consortium that allows others in that consortium access. I don't know the fees or what not. But the consortium members must have a chinese presence (I'm pretty sure they have to be Chinese owned, but I'm not sure), and TD-SCDMA research is funded by the Chinese gov't and they are trying to push it over the other standards. That's alot of market pressure, and I don't think the Chinese gov't would allow its companies to do protectionist, and monopolisitc practices as the american company does to other chinese companies in the consortium. Other companies are fair ball of course.
In the future you might find that TD-SCDMA will come to the front because the technology is cheeper. I have no idea if it will be better. But it will come from China where labour and research is cheep compared to the rest of the world.
$30,000 - $50,000 is about what it's going to cost to employ another full-time programmer in a developed country.
Salary counts for less than half of the total costs of having a salaried employee. There are taxes, benefits, building costs, equipment costs, training, etc.
Do you really think that they can get good technical employees who are willing to work for US minimum wage (and probably less than minimum wage in France)?
To give a good example of this. Back when I was well paid, and I use to deal with engineering burden rates for figuring out the cost of my time in dollars for upper management, I had a set figure, that was $625 US a man day. This was for the canadian office.
If you took the average salary at the canadian office you'd see it work out to something like this:
Engineering Cost per year $162,500
Engineering Salary per year 45,360
All the rest: 117,410 (72%)
A dozen engineers would cost: $1,950,000
By the grand parents numbers of advertising (which I don't have on my screen so I can't see them exactly Damn I think its 30k per year per advertizer) Mandrake needs 65 advertisers just to handle the 12 engineers.
Of course part of the burden (appart from the engineer's salary) is the salary for 'non-production' personal, or non-billing personal. This is all your C*Os, secretaries, and marketing. It does not include production costs, those are usually covered under a different system.
But welcome to the world of 'too much information, more then I wanted to know' world of project management and budgeting.
Simpsons' Quote:
Willi hear you, Willi don't care.
>
Why is it that there isn't a push one way or the other to amend the constitution to clarify the "right to bear arms"?
Being a Canadian I always thought this was a better interpretation. You have all the right to go out into the woods with a pen knife and get all the bear arms you like.
Or just interpretate it to mean you're allowed to wear short sleeves. But this would make all the $5 T-shirt producers, who put something on them and sell them for $25, happy.
So, hollywierd is upset that word of mouth is traveling faster than ever. Why don't they start making movies that don't suck? As for EULAs. I didn't sign anything. I didn't agree to anything. They can put them up there if they want to.
Then they'll start forcing you to pay by credit card. And on the little bill right by the signature portion they'll put in super fine print... "By signing this you agree to abide by the EULA"
Of course for them to do this they'll have to find a way to get people to stop paying cash. All they have to do is up the price a bit more. Its already getting to the point where I don't go to the movies because its too expensive and when I do go I'm hesitant giving over cash because I don't carry that much with me anymore.
They could also weasle it in as 'Oh won't someone think of the children' by making R rated movies require the credit card as proof you're an adult. A driver's license won't do because movie theatres will probably get rid of cashiers to look at them, and they won't be hooked into the licensing network for those of us who have magnetic strips on their DLs.
But this is just getting cynical.
This collapse was predicted years and years ago. When I was working for a power equipment manufacturer (transformers, relays, switches, capacitors), marketing was playing this sort of thing up to the utilities.
As equipment gets old it becomes less and less efficient. This includes the transformers that bring the power from high voltages to low voltages to your home, and the generators that produce the power. AND it includes all those Air conditioners that are running in hot weather.
No-one has been able to afford to bring new generators online recently. And probably not to upgrade/replace old less efficient equipment. And I'm sure most people haven't bought new ACs either because of the economy.
It wasn't so long ago that something similar happened to Chicago during a heat wave there. And we all remember a few years ago that California had rolling blackouts because the grid couldn't handle the power. And NYC suffered similar blackouts for the same reason in the 60s and 70s I believe.
Unfortunately since its such a large grid its going to take a while for it to come back up. You have lots and lots of main power generators. Each one has to be brought back onto the grid one at a time. Each one has to be synced to the current AC 3 phase system within 5-10 degrees of what's there or when that generator comes on it might cause all the generators to drop out. Syncing a generator takes time and patience.
Then you have to bring the consumers back on. Every time you bring a new section on you have a hell of alot of inrush current as Air Conditioners and motors start up. This is why your lights dim a bit when you turn on certain pieces of equipment. Imagine the dimming you get as 1000 Air conditioners come on at once. If its too much a relay might trip off and the grid might collapse under the strain as a generator falls offline. And yes this is a real meaning to the word offline, the generator is not on the power lines anymore.
It will take time for everything to come back up, and con-ed isn't going to rush it. They will take it up in stages, make sure that everything is ready to go before bringing up the next stage. A collapse this large can happen again and again if they rush. But it might be quicker, the reports don't saay how big the failure was and how many generators fell offline. It could just be that all the distribution substations tripped, but I doubt it. For this to be so widespread the generator protection relays probably all fired off and took their generators off the power grid.
Don't you just love cascading failures? Overloaded power grid; all the generators are close to their shut off point. One fails, all generators go into the range of shutoff, and off they go one right after the other. They probably all fell offline withing 30 seconds, and will probably take 3 days to come back on fully.
Sounds like your power supply can't handle the inrush current for the start up of your hard drives and fans.
When motors start up they have to get over the standing still friction of the unit before they get over the hump of startup and then into the lower friction of a moving unit. I would put in a nice graph but slashdot won't let me.
What the graph looks like is there's a rise up to a peak. Doing an integration under the line to the peak gives you the force needed to get to the peak. This is the force needed to get to the object moving. The graph then falls down a fair ways, (maybe almost to zero it depends on the two materials being being rubbed together), and then the line stays steady, horizontally. This represents the force needed to keep the object moving.
This is about all I remember from College physics on friction. There's also some other things that require a high inrush current during startup. But motors are the biggest thing.
So when you rock your power switch just right you leave the motors running enough not to require the inrush current. Shut off power to the motherboard to force it to restart, but you don't leave the power off long enough to cause the power supply to go into suspend mode and turn off all the power to the system.
Probably the best thing to do is get a power supply that handles a larger inrush current then the current one. But there could be something else wrong with the system too, and I haven't built a system for close to 8 years and my knowledge might be a bit dated.
"those who can, do. Those who can't, teach." :p
What about those who can teach?
Well those that can teach, teach. Those who can't teach, teach teaching.
Now say that 3 times fast, I dare you.
According to the latest GamePro, the "CD-only version of Sony's own system [was] internally named 'Play Station X' (which is why you sometimes see the abbreviation 'PSX' for original PlayStation stuff)." So basically, that's all there is to the PSX name. As far as PS1 and PS2, I'd imagine it's just so you can tell the original playstation apart from the revamped one and from the PS2.
You won't believe how often X makes it into marketing names because sometimes marketing people have no idea on names. You'd think they would, but they don't.
I use to work for a company that named their products Form [number] or something like that. With the initials F1, F2 and so on through the generations. It wasn't a very imaginative name to begin with. There was the F3 on the market and the engineers started work on the new product to be called the F4. But before that product hit the market or even started marketing it there was a need for an upgraded F3. So the engineers had no idea what to call it, lots of humming and hawing went around the place. The drafters got fed up with not knowing what to title their diagrams and put it into the system as FX, fully intending to go back and fix it later when the marketing department came up with a name. They different, and that company still makes the FX, even though they make the F4, F5 and now the F6.
The PSX name probably came out for the same reason. Sometimes Marketers can't come up with names, and give up too easily and just use the internal names. Which are usually confusing and unsuitable for the mass market. I'm not saying coming up with a suitable name is easy, but that's what the marketing department gets payed for.
I don't think there can be an appeal on an innocent verdict in a criminal trial in the states.
I think its called the double bind clause in their bill of rights. If you're charged for murder and get found innocent you can't be brought to trial again just because they found some new evidence.
I know in Canada that an innocent verdict can be appealed. And in non-criminal cases the double bind clause doesn't apply. But this was a criminal copyright trial wasn't it?
That's not as far off base as you'd think. Crientien (or how ever you spell his name) signed an agreement with Bush about mutual protection or whatver between US and Canada.
Basically it states that the US can invade Canada if there is an emergency. The agreement leaves out who declares the emergency (Bust probably), or what the emergency is (Canada is selling energy for a profit, probably), (Or there's a pirate in Inovac, and the only way to get there is to go through the rest of Canada).
As a Canadian, this just sickens me. But then again since when did anyone listen to the voters.
There is a hard limit on 2G CDMA of 61 cell phones per base station. But the caculations I was mentioning before (damn wish I still knew where they were), assumed a Frame Error rate of 2%. That still provides a reasonable voice quality rate. 1xRTT (the 3G that sprint and verison are using) has the same limitations on codes to use for cellphones. (That's not entirely correct, but I doubt that sprint and verison are using the more advanced 3G features that allow them to boost capacity without boosting bandwidth. I don't think most customers phones support these features yet)
There are also a whole bunch of other issues, dealing with what vocoder is being used, which also determine both voice quality and the ability to do error correction. The '8k' vocoders have better error correction then the '13k' vocoders but they sound like shit.
I don't know if Verison or Sprint are doing the same thing as Telus is doing. But Telus defaults to the 8k vocoders for all calls, except the calls to their support lines that are at 13k. Hmmm I wonder why? (cynism)
From the sounds of things both Sprint and Verison are running at maybe 5% FER and maybe 55-60 calls per base station. You'd have to look at the signalling traffic to be sure.
CDMA is also generally reguarded as more advanced outside of the US as well. Which is why the air access portion of CDMA is the basis for most 3G standards. Going forward GSM providors are moving towards UMTS (or is it another acronym starting with U, I don't remember). And that's basically another name for W-CDMA.
Its a bit of a mess of why there are divergent paths. But the big one, is that Qualcomm owns most of the patents on CDMA, and are really the only player for CDMA base band chips. I only know of one other company that has had their chips certified by Nortel and Lucient to work off their base stations, and they're just getting into production now.
GSM was out in the market first. Europe had alot of problems with 1G (analog) cell phones. There were so many different standards that you had the same problem moving form one system to another, that you have in the US. When 2G came along GSM was choosen for the most part to hold back these woes. I don't remember fully but I think GSM came out in the mid 80s. Qualcomm introduced CDMA in the early to mid 90s, about 10 years behind GSM. So GSM had a long time to get entrenched.
But GSM is alot like TDMA, at least at the physical layer. CDMA is completely different. Basically TDMA has the same limit as AMPS does. You can really only have one transmitter on one frequency at a given time. CDMA changes this, allowing multiple transmitters to be on all the time on the same frequency. Theroitically this means that CDMA is interferance limited to the number of cellphones talking to a base station, while TDMA/GSM is still frequency limited. I don't remember the exact numbers, but it works out to around 8 GSM phones can effectively share the same frequency, while 43-47 CDMA phones can share the same amount of frequency. Its a complicated formula, because CDMA uses more spectrum per signal, but you can have base stations sitting almost ontop of each other physically transmitting on the same frequency. These numbers are for 2G CDMA. 3G CDMA is suppose to increase capacity by 1.7 times.
Of course Europeans have a burr up their saddles when it comes to technology from the States (and the US has the same burr it seems). And Qualcomm isn't helping manners. Qualcomm is greeder then Microsoft, and are probably impedding the adoption of CDMA more then anything else because of their monopolistic attitude (Think threats to customers who consider using other chip sets, exhorbadant license fees, etc). True 3G CDMA does meet all the requirements of high speed and higher capactity. Sprint and Verison are not deploying true 3G CDMA. True 3G CDMA has a bandwidth an effective bandwidth of 5 MHz, just like W-CDMA has. Instead S&V are deploying 1xRTT which is backwards compatable with 2G cellphones. Don't ask me why CDMA-2000 and W-CDMA are not compatible with each other, its all infighting between the GSM manufacturers and Qualcomm. The basis of it is that if CDMA-2000 was the 3G standard, and backwards compatible with 2G CDMA, then why would new carriers want to buy GSM equipment if they're going to have to get rid of them later, why not just buy 2G CDMA and upgrade to 3G later.
But this is the cellular market place. If they can get more money out of their customers for less service they will. And this sort of attitude isn't at the provider/network level but basically goes all the way up to the base station manufacturers who run the international standards committees.