it's simply a matter of aiming a strong signal at the uplink transponder on the satellite
Worse, send a really powerful signal (read- military radar magnetron hooked up to mondo dish) and you can permanently fry that transponder, and do so with a burst so brief and directed that it's not terribly easy for anyone to figure out whonunnit. It's a great piece of asymetric information warfare - spend a couple of million dollars to knock out a few dozen civilian comsats, each tens or hundreds of millions of dollars worth. Best of all, it's quite possibly not in breach of any international treaty !
Excuse me, I have to go now, my mechanical pirnahas are hungry...
I'm sorry, IANAMT (I am not a military tactician) but do we really need to be able to strike someone within a period of two hours?
Moreover, do you want a bomber that fast? I figure a nice big slow subsonic B52 chugging sluggishly over the pole gives plenty of time for cooler heads to prevail, rogue commanders to be unseated, mad dictators deposed, invading armies halted, and someone to find the power cord for the WOPR.
Hypersonic jets are cool and all, but I think they would be better applied in the commercial market
The same could be said for supersonics, and indeed was for Concorde. The trend for commercial aviation appears to be in the opposite direction, however. At least in Europe and North America, everyone seems to want the lowest cost, maximum discomfort solution (SWA, JetBlue, EasyJET, Ryanair, GermanWings, etc.).
On an unrelated note which no-one seems to have mentioned today: the last time such a TAV (trans-atmospheric-vehicle) was proposed officially (orient-express/hotol) everyone got twisted over the damage it would apparently do to the ozone layer (it was said that the hypersonic TAV would fly in the ozone layer and that its wake turbulence would disrupt the layer, producing a big 'cut' in it).
If customers assert their right to control something they (gasp) actually own, they might (gasp) get ideas above their station.
Maybe they'll want to show their kids a version with sponsorship messages and product placements removed. It's not difficult to imagine a (PC based) player that takes a "blurtrack" file which matches a DVD, and superimposes a blur over parts of the screen that I don't want crammed down my throat.
Maybe they'll want to watch the basketball but have the TV show a replay rather than listen to the network's shamless shill proclaim "I'm going to Disnaeland".
Moreover, the EFF is defending the principle that the customer should control what they've already paid for. That the customer can watch a US region movie in Australia. That the customer who bought the home version of "I know what you did last Tuesday" can watch it on their laptop, on their cellphone, can listen to the soundtrack without the dialog, can skip over the ten minutes of trailers and ads that preceed it.
Hollywood doesn't want the consumer having this control. It devalues their advertising and prevents them from reselling you the same material over again in each format you want to use.
Everyone concentrates on 3G's bandwidth as being its predominant value-add, and it's difficult at the moment for the phone companies to figure out a sufficiently compelling application for all that bandwidth. But 3G has some other features which make it interesting - it's a shame the CDMA and GSM markets didn't include these in their 2G/2.5G offerings.
Chief among these are:
Packet-switched operation. To transmit data (except SMS messages) it's necessary to open an end to end virtual circuit. So you can't trickly information back and forward to the phone all the time, at a very low bandwidth (and consequently very low cost). And there's no multicast, so software download to each phone has to be done one at a time.
Location-specific services. "Where am i?", "Where is the nearest gas station?", or that DoCoMo fave "beep me when a single girl my age who also likes ninja manga is nearby".
These don't need 3G's bandwidth, but the 2G network can't really deliver either. If the phone companies had been conservative and added the above, they'd be in clover. That's not just 20-20 hindsight - DoCoMo in Japan did both, and they're making enough money to actually pay for the 3G network they're building - and simultaneously getting their consumer base onboard with the idea of getting games, media, etc. on their phone.
Reiser FS is already a pretty mature, stable, usable product. Once V4 is done, is there really much work left to be done on ReiserFS proper? Do you have a giant to-do list that'll keep you and the guys occupied for years, or do you intent to work in a diffent direction (SAN, networkFS, databases, etc.)?
(or perhaps you'll just retire to Portugal and play lots and lots of golf)
The USB cable has an upside of being able to charge my phone while connected to my computer. But then again, I can only use this $40 cable for my phone. Since my phone is bluetooth
I had a similar choice (heck, without that nice USB option) on my Nokia phone. It felt a _lot_ like the Bluetooth mafia making me an offer I couldn't refuse. That $40 for the USB cable in criminal. I bet it's not a USB-OTG connector, but some hideous proprietary one.. A standard cable would cost what, $5? As a big Bluetooth IP owner/vendor, Ericsson has an agenda of its own, which doesn't necessarily mean always giving you the best choices.
So that's Ericsson's plan - they priced the cable HIGHER than the Bluetooth connector so they could strongarm you into being a Bluetooth adopter, and made it a proprietary connector so they couldn't be undersold by a generic supplier.
If bluetooth could stand on its own technical merits then they wouldn't have to resort to such gangsterism.
Remember, 802.11, like ethernet, is a lower-level protocol than TCP/IP
Ah, it's nice to see someone was paying attention when the network-stack diagrams were being handed out.
... protocols such as Appletalk or DECnet as well. Thus, one could very easily implement a protocol that does everything you just described, yet still have essentially invisible expandability up to talking with a "real" network.
Quite. Appletalk, qnet, and SMB/CIFS have been doing fairly-lightweight network/device dynamic discovery for years on ethernet and other carriers (and really, as a transport bluetooth is way more complicated than 10-base-T ethernet).
Re:A wired standard would be more useful to me.
on
The Death of Bluetooth?
·
· Score: 3, Interesting
USB does everything you ask, including the power, very well. Just about the only thing that it lacks, and that your applications above really call for, is a lightweight connector. If they standardised a little clip-in edge connector (like the ones that PCMCIA modems use to connect to the phoneline-adaptor dongle thing) for stuff like your earphones, then you'd be set.
For all I know, the USB folks do have such a standard mini-connector.
bluetooth vs 802.11 - a vendor's opinion
on
The Death of Bluetooth?
·
· Score: 4, Informative
A few years ago (probably three, I think) I did some business with a company who made both solutions and complete products for both proprietary wireless systems and the then-new 802.11 and bluetooth markets (guessing which is left as an exercise). I asked their tech-sales guy which of the latter pair he thought would win, or would they both succeed - "only 802.11" he replied, "without a doubt".
I responded (as have many slashdotters above) that surely the two weren't for the same task, and thus were surely destined to exist in parallel, in adjacent market sectors. He told me this wouldn't be true, and his explanation went something like this:
Bluetooth has two main selling points:
it's a far simpler protocol, which means a smaller, cheaper digital side (fewer, smaller, slower, cooler, lower-power parts)
it's lower powered, which means a cheaper analog side (and less power consumption, and thus longer battery life)
But, he predicted, both of these would be eroded quickly. The former would vanish, he said, when both bluetooth and 802.11 are cost-reduced down to single-chip solutions (which now they mostly are). Sure, the 802.11 chip is bigger than its bluetooth buddy, but the cost-differential is pretty slight.
The latter would still apply, but he predicted (and it's come true, although not yet productised) that the 802.11 folks would produce a low-power, short-range version.
So one of Bluetooth's advantages (for its own market) is largely obviated, the latter partially so. Set this against the economies of scale that 802.11 enjoys, and the greatly enhanced oppertunities for interoperation that the dominant standard enjoys, and the "roaming" use of Bluetooth is beaten, resoundingly.
Bluetooth had two other markets in mind:
as a desktop "wire replacement", for keyboards, mice, etc., essentially replacing the many proprietary protocols in that space, and for stuff for which we now use USB. Bluetooth can't beat USB on cost or performance, which leaves only the mouse and keyboard market. Sure, it would be nice if your digital camera or mp3 player would just "see" the PC without your having to hook up some cables, but the cable burden isn't that onerous. Now, 802.11 may be cheap, but I think it'll be a few years yet before I have an 802.11 mouse. So there is a market there for bluetooth, but it's small, and bluetooth isn't compelling enough to displace the proprietary guys there.
the true "personal area network", where all the devices I wear interact with one another, and dock to my pc when I sit at my desk. Well, all those devices (PDAs, cameras, cellphones, mp3) have (or are now) converging on a single device (which we call a phone, 'though it is many things besides). This leaves bluetooth with another thin market segment - connecting cellphones to PCs. Again, cheap fast cables, IRDA, and proprietary RF solutions all own this space, and again bluetooth isn't compelling enough to displace them.
Every product has to answer the same challenge - not "are you useful" but "are you useful ENOUGH".
Did the NullSoft buyout contract specify that they had to keep them on for a decade?
Very possibly ('though probably four or five years, not a decade) - buyout contracts often do, to prevent the "human capital" from taking their stock and running. The carrot to folks is that they get lots of new options, which vest annually so long as they remain.
Once the deal is signed, both sides often try their best to wiggle out. The stock options aren't paid out if the employee quits early, so the company tries to get the employee to quit. CEOs become directors of empty divisions with no staff and no mission, stuff like that. The company can't be _too_ blantant about it (i.e. make the CEO unblock toilets all day) as that's constructive dismissal, in which case the employee can leave with the stock (after lots of legal squabbling, of course). Equally, mr small-company-entrepreneur type wants to get the stock and bug out (either to his next startup or to Hawaii) and doesn't want to be a drone for the next half decade. So he _tries_ to get constructively dismissed. Fired for gross misconduct (not showing up, punching out his boss, etc.) won't work - so he has a bad attitude, doesn't bathe, says dumb things to the media, produces product that makes his employer uncomfortable, founds the aryian-spaceship-league, whatever. So a war of attrition is fought.
Naturally, I don't know the terms of the nullsoft acquisition, but it may be this is Frankel's (et al) idea (or at least in his mind). I figured this was the case when Gnutella came out (AOL were _never_ going to be happy with that) and WASTE is even more AOL-unfriendly (heck, it's got a chat client - who needs AIM?).
Someone should write a book about the constructive dismissal stories that fill Silicon Valley - Sculley sending Jobs to his own office building to do nothing (Jobs cracked rather quickly). I heard of some guy coming to work dressed in a full frogman suit (including flippers and mask) and walking down in the corridor when customers were around - company dress code said "no shorts, wear shoes" - if they'd changed it to read "no bodyglove swimming attire" just for him, then that would have been the constructive dismissal he sought.
If Sony were the monolith its detractors claim, the PS developers would be pretty much mandated to run ITRON/uITRON/JTRON in this kind of device. I think Sony licences an emotion engline uITRON build to games developers who want one - does anyone know for sure what RTOSes do work on EE?
The light and mirrors solution is, from a safety perspective, greatly superior to a transparent screen. Existing projection-reflected-from-windshield systems use lenses to project the image focussed at infinity. That way a driver (whose eyes are already focussed at infinity when looking at the road ahead) doesn't have to pull focus in to 1m to read the HUD and then push focus back out to infinity to safely read the road ahead. Most of the time spent looking at your speedo on a normal car is the focus shift time, rather than the eye-movement time, so a projected HUD like the GM one you describe is the best solution. The HUD reticles used by military pilots also use projectors, and similarly project the pitch-ladder and other indicators at infinity, so they appear to float spectrally "out there" in front of the aircraft.
For regular computer display purposes, a transparent screen doesn't seem terribly useful, due to contrast and "visual noise" interference from whatever is behind the screen (mitigated a bit if the screen is frosted).
Still, there's plenty of possible applications for this:
Just make a graphics card with more transistors and drop the traditional processor
There's a lot of work being done on reconfigurable computing, which imagines replacing the CPU, GPU, DSP, soundcard, etc., with a single reconfigurable gate array (like an RAM-FPGA). You'd probably
have a small control processor that manages the main array. On this array one could build a CPU (or several) of whatever ISA you needed, and GPU, DSP, whatever functionality was called for by the program(s) you're running at the current moment. Shutdown UnrealTournament 2009 and open Mathlab, and DynamicLinux will wipe out its shader code and vector pipelines, and grow a bunch of FP units instead. Run MAME and it will install appropriate CPUs and other hardware.
In the initial case, this would be controlled statically, a bit like the way a current OS's VM manages physical and virtual memory. Later, specialist "hardware" could be created, compiled, and optimised, based on an examination of how the program actually runs (a bit like a java dynamic compiler). So rather than running SETI-at-home your system would have built a specialist seti-ASIC on its main array. There will be lots of applications where most of the work is done in such a soft ASIC, and only a small proportion is done on a (commensuately puny) soft-CPU.
This all sounds too cool to be true, and at the moment it is. Existing programmable gate hardware is very expensive, of limited size (maybe enough to hold a 386?), runs crazy hot, and doesn't run nearly quickly enough.
I hope they keep the building (although if they're not using it, I suspect they won't).
It adds another weird element to the already surreal aspect that Ames/Moffet presents, particularly to the north. There's a number of odd (nay, sinister) looking buildings, some positively Quatermassey domes, weird towers, and of course the giant rectangular intake of the wind-tunnel building. The whole place has a cool area 51 big science of the 60s feel about it.
Combine that with the Mountain View city lot beside it, where they keep hundreds of trees and bushes in wooden boxes, ready to be transplanted, lined up in neat little rows - it looks a bit like the set of The Prisoner.
Nearby is SGI's main campus, where they've build a couple of ultra-modern office buildings (not as short of cash as we may have thought). Given that SGI's major remaining customers are NASA and NSA, it's get another little piece of the "look what government money built" zone up by Shoreline.
IMNAMD (I am not a marketing dweeb) but I have to ask - what's their (Rocklyte's) value proposition?
There are several free, reasonably mature windowing environments available for linux
already, many featuring hardware acceleration. Several are suitable for embedded
use. Why do I want to spend $40 for this? (I'm not being rhetorical - the site
isn't accessable). There are innumerable linux distributions, several of which
boot straight from CD without install.
Frankly, the speed differentials Scitech quotes (over Xfree) aren't really all that
impressive for most graphics adaptors. Sure, there's a big difference between unaccelerated (e.g. vesa) access and
accelerated, but a 20% differential between the 2D performance of one accelerated solution and
another just isn't that compelling. Now many applications are _that_ dependant on
2D performance? If I'm that 2D bound I can spend that $40 and get vastly better
graphics performance by buying a better card.
The "foo is old fashioned", "foo is too complicated", "foo is SO last century" claims
that some make (I dunno if these guys do, as their site is still down) aren't value
propositions. Is something significantly faster? Significantly smaller? Significantly
more useful features? Significantly cheaper? Those are.
Parenthetically, note that I don't apply this standard to free projects. Someone can go
code a new OS just for their own pleasure, and doesn't have to pass a customer-value-proposition test. Why? Cos they don't have customers, and so they're not obligated to provide value to anyone.
My job requires me to develop for QNX using their "momentics professional" kit, hosted on windows. QNX chose to protect this with FlexLM, which is in turn protected by some of the Macrovision stuff that Intuit use.
While the QNX stuff is generally of excellent quality the FlexLM thing is a persistent source of problems. Installation and upgrade have never gone smoothly, with obscure services not starting or being misconfigured by the installer, client authentication going wrong more times than it should (i.e. ever), and occasional file-locking problems that require a reboot. At least in my case, licence management seems to generate as much traffic with the QNX support folks as does their actual product (host and target) in its entirety.
Worryingly, the licence is bound specifically to one licence server. I _imagine_ that if the machine (a laptop) were destroyed, lost, or updated, then there would be some means whereby I could persuade them to issue another licence, but it's bound to be a sticky point.
I wouldn't care if everything worked properly and transparently, but it doesn't. My vendor is essentially treating me like a thief and simultaneously making himself look like a bozo (which he isn't - the rest of the QNX stuff is great).
On the last occasion it took several days to resolve the licence manager issues - had this been at a more critical time then this would have been a dealbreaker. It leaves me with a rather bitter taste in my mouth, and I'd think twice before recommending QNX to another client, purely for this reason.
So is their bizantine DRM saving them money, or costing them? I think Intuit can answer that for them.
Corporations generally don't use their own legal departments to engage in litigation. The corporate legal folks are used for contracts, NDAs, employment law, intellectual property (patents, trademarks, copyright), government relations, and compliance (SEC, FTC, EPA, etc.).
Either to engage in, or to defend, litigation a large corporation would generally engage a law firm specialising in litigation, and probably one specialising in the particular type of litigation - employment, competition, IP, environmental, etc.
So the question should be "how big is apple's legal budget?".
The Sun and IBM java folks are, informally, intertwined in a number of ways:
The Sun java folks used to (until last year) work from two buildings on DeAnza in Cupertino. At least one of these was an ex-Taligent building, and consequently IBM owned the furniture (I think the lease may also have been some kind of sublease thing). One time IBM wanted all their furniture back, and I
believe they flat refused to sell it, forcing each Sun java employee to move out of his office into the corridor, while the facilities dudes came and swapped his desk etc. out for an essentially identical replacement.
The sun java folks are now confined largely to Sun's Agnew's development centre, built on the site of the county mental hospital. Given that Cupertino was a totally excellent place to work, and Santa Clara most assuredly isn't, I'd guess that if the IBM folks said "we'll buy java, and y'all can come back and work in Cupertino" there would be a lot of happy people.
One of IBM's largest Java development centres is (waitforit) on DeAnza in Cupertino, right beside the old Sun java building. Both are former Apple buildings, and a bunch of the java folks are ex-apple.
I wouldn't put too much stead in the "disgruntled employees veto the deal" theory, mentioned above. These days, the average Silicon Valley employee cares about 1) do I get paid ? 2) does my commute get better or worse ? 3) do I get to do something that isn't totally crap ? (the former number 1, "will my stock options make me rich?" no longer figures much).
I remember reading a thing about Adm. Hyman Rickover, the father of the US nuclear submarine program. He was considering the means by which motive power would be extracted from the nuclear reactors. The scheme had turbine blades inside the reactor vessel, turning a shaft that (eventually) turned the boat's screws. This mean the shaft had to pass through the wall of the reactor vessel. He was worried that the seals around this opening wouldn't be perfectly reliable, and naturally if they ever failed this would allow radioactive fluid into the boat's compartments, irradiating the crew. The seal manufacturers assured him they could make a seal that was perfect, that would withstand all that could be thrown at it. Rickover wasn't sure, wondering if a magnetic interlock (where the reactor
vessel is intact, and magnets on either side cause one shaft to move its counterpart).
Rickover took the seal guys aside, and asked them - if your son was on this boat, would you still want seals, or would you opt for the magnetic method? The seal guys thought for a while, and sheepishly
replied that they'd go with the magnets. To this day, all US naval reactors have magnetic interlocks, not seals.
Fact is - seals are hard. Hard to make, hard to maintain, and hard to check. They're almost always the first thing to fail, and rarely gracefully.
So, rather than the next gen spaceplane being some slicko streamline hitech composite fibre whatnot, it should be a windowless monocoque made from thick polymerised concrete. The astronauts will need a stihlsaw to go EVA, but then a concrete spaceship needs no maintainance, so they won't have to.
Those large projects that move forward at a decent pace seem to be those that have a high tolerance for forks. Forks are generally "considered harmful", but in fact forks in a tolerant, open-minded, and "adult" environment are highly beneficial.
Good forks have the following in common:
they fork off to do major changes to an existing product, changes that require a
destabilisation of the codebase that would prevent the main product from doing
necessary maintainance and incremental fixes
the "factions" (the forkers and the forked-from) stay on good terms. Everyone keeps
a (mostly) level head, and both factions see the wellbeing of the other as important.
changes from the fork are migrated back, piecemeal or wholesale
often either the fork or the original branch are deprecated, and the fork fused
Consider some good forks:
mozilla -> phoenix -> mozilla(whateveritisbird)
X11 -> XFree
GCC2 -> EGCS -> GCC3
linux is perhaps the best example - two major branches running all the time, and both
(particularly the 2.3, 2.5, etc. dev fork) heavily forked themselves. 2.5 changes are
often backported to 2.4, even to 2.2, and the maintainers all still talk to one another.
By way of contrast, the GNUemacs/Xemacs fork is a prime example of a bad fork. Bad blood,
wilful incompatibility, divergence, duplication of effort.
If XFree's current "governance fork" turns into an all out code fork then that would, I fear,
be a bad fork - all that bad blood will surely make things very difficult technically.
So perhaps the best advice to a successful project is "encourage forks, and provide a
safe environment for them". Apache and Mozilla both do this, to their benefit and credit.
On a related note, can anyone recommend a decent open source / free software graphical font design tool ? I looked into this a few years ago and things deemed to be in a crufty state of disarray. What do folks use now?
I see from the article that the trunk connectivity comes from an undersea fiberoptic cable. I think I remember (perhaps from that Neal Stephenson WIRED article?) about a fat copper cable that folks were putting around Africa, on land, which I think was called Ring Around Africa. I think I also remember there being problems with folks stealing parts of it (as it represented a decent value for copper recycling) and risking their lives due to the current.
Does anyone remember this (or is it just my imagination), and if so, what became of it?
Slashdot Mirror
Worse, send a really powerful signal (read- military radar magnetron hooked up to mondo dish) and you can permanently fry that transponder, and do so with a burst so brief and directed that it's not terribly easy for anyone to figure out whonunnit. It's a great piece of asymetric information warfare - spend a couple of million dollars to knock out a few dozen civilian comsats, each tens or hundreds of millions of dollars worth. Best of all, it's quite possibly not in breach of any international treaty !
Excuse me, I have to go now, my mechanical pirnahas are hungry...
Moreover, do you want a bomber that fast? I figure a nice big slow subsonic B52 chugging sluggishly over the pole gives plenty of time for cooler heads to prevail, rogue commanders to be unseated, mad dictators deposed, invading armies halted, and someone to find the power cord for the WOPR.
Hypersonic jets are cool and all, but I think they would be better applied in the commercial market
The same could be said for supersonics, and indeed was for Concorde. The trend for commercial aviation appears to be in the opposite direction, however. At least in Europe and North America, everyone seems to want the lowest cost, maximum discomfort solution (SWA, JetBlue, EasyJET, Ryanair, GermanWings, etc.).
On an unrelated note which no-one seems to have mentioned today: the last time such a TAV (trans-atmospheric-vehicle) was proposed officially (orient-express/hotol) everyone got twisted over the damage it would apparently do to the ozone layer (it was said that the hypersonic TAV would fly in the ozone layer and that its wake turbulence would disrupt the layer, producing a big 'cut' in it).
Maybe they'll want to show their kids a version with sponsorship messages and product placements removed. It's not difficult to imagine a (PC based) player that takes a "blurtrack" file which matches a DVD, and superimposes a blur over parts of the screen that I don't want crammed down my throat.
Maybe they'll want to watch the basketball but have the TV show a replay rather than listen to the network's shamless shill proclaim "I'm going to Disnaeland".
Moreover, the EFF is defending the principle that the customer should control what they've already paid for. That the customer can watch a US region movie in Australia. That the customer who bought the home version of "I know what you did last Tuesday" can watch it on their laptop, on their cellphone, can listen to the soundtrack without the dialog, can skip over the ten minutes of trailers and ads that preceed it.
Hollywood doesn't want the consumer having this control. It devalues their advertising and prevents them from reselling you the same material over again in each format you want to use.
M-x show-mappings
C-M-g pain
C-M-G agony
C-M-T paralysis
Look at the inventors of *emacs: Stallman - CTS. Gosling - CTS. Zawinsky - weird. Wing - bald.
In the absence of "emacs peddles", the confirmed emacs user is doomed (dooomed).
(in next week's exciting episode: "Perl and your spermcount - the shocking truth").
Chief among these are:
- Packet-switched operation. To transmit data (except SMS messages) it's necessary to open an end to end virtual circuit. So you can't trickly information back and forward to the phone all the time, at a very low bandwidth (and consequently very low cost). And there's no multicast, so software download to each phone has to be done one at a time.
- Location-specific services. "Where am i?", "Where is the nearest gas station?", or that DoCoMo fave "beep me when a single girl my age who also likes ninja manga is nearby".
These don't need 3G's bandwidth, but the 2G network can't really deliver either. If the phone companies had been conservative and added the above, they'd be in clover. That's not just 20-20 hindsight - DoCoMo in Japan did both, and they're making enough money to actually pay for the 3G network they're building - and simultaneously getting their consumer base onboard with the idea of getting games, media, etc. on their phone.Reiser FS is already a pretty mature, stable, usable product. Once V4 is done, is there really much work left to be done on ReiserFS proper? Do you have a giant to-do list that'll keep you and the guys occupied for years, or do you intent to work in a diffent direction (SAN, networkFS, databases, etc.)?
(or perhaps you'll just retire to Portugal and play lots and lots of golf)
I had a similar choice (heck, without that nice USB option) on my Nokia phone. It felt a _lot_ like the Bluetooth mafia making me an offer I couldn't refuse. That $40 for the USB cable in criminal. I bet it's not a USB-OTG connector, but some hideous proprietary one.. A standard cable would cost what, $5? As a big Bluetooth IP owner/vendor, Ericsson has an agenda of its own, which doesn't necessarily mean always giving you the best choices.
So that's Ericsson's plan - they priced the cable HIGHER than the Bluetooth connector so they could strongarm you into being a Bluetooth adopter, and made it a proprietary connector so they couldn't be undersold by a generic supplier.
If bluetooth could stand on its own technical merits then they wouldn't have to resort to such gangsterism.
Ah, it's nice to see someone was paying attention when the network-stack diagrams were being handed out.
Quite. Appletalk, qnet, and SMB/CIFS have been doing fairly-lightweight network/device dynamic discovery for years on ethernet and other carriers (and really, as a transport bluetooth is way more complicated than 10-base-T ethernet).
here and here and here.
For all I know, the USB folks do have such a standard mini-connector.
I responded (as have many slashdotters above) that surely the two weren't for the same task, and thus were surely destined to exist in parallel, in adjacent market sectors. He told me this wouldn't be true, and his explanation went something like this:
Bluetooth has two main selling points:
But, he predicted, both of these would be eroded quickly. The former would vanish, he said, when both bluetooth and 802.11 are cost-reduced down to single-chip solutions (which now they mostly are). Sure, the 802.11 chip is bigger than its bluetooth buddy, but the cost-differential is pretty slight.
The latter would still apply, but he predicted (and it's come true, although not yet productised) that the 802.11 folks would produce a low-power, short-range version.
So one of Bluetooth's advantages (for its own market) is largely obviated, the latter partially so. Set this against the economies of scale that 802.11 enjoys, and the greatly enhanced oppertunities for interoperation that the dominant standard enjoys, and the "roaming" use of Bluetooth is beaten, resoundingly.
Bluetooth had two other markets in mind:
- as a desktop "wire replacement", for keyboards, mice, etc., essentially replacing the many proprietary protocols in that space, and for stuff for which we now use USB. Bluetooth can't beat USB on cost or performance, which leaves only the mouse and keyboard market. Sure, it would be nice if your digital camera or mp3 player would just "see" the PC without your having to hook up some cables, but the cable burden isn't that onerous. Now, 802.11 may be cheap, but I think it'll be a few years yet before I have an 802.11 mouse. So there is a market there for bluetooth, but it's small, and bluetooth isn't compelling enough to displace the proprietary guys there.
- the true "personal area network", where all the devices I wear interact with one another, and dock to my pc when I sit at my desk. Well, all those devices (PDAs, cameras, cellphones, mp3) have (or are now) converging on a single device (which we call a phone, 'though it is many things besides). This leaves bluetooth with another thin market segment - connecting cellphones to PCs. Again, cheap fast cables, IRDA, and proprietary RF solutions all own this space, and again bluetooth isn't compelling enough to displace them.
Every product has to answer the same challenge - not "are you useful" but "are you useful ENOUGH".Very possibly ('though probably four or five years, not a decade) - buyout contracts often do, to prevent the "human capital" from taking their stock and running. The carrot to folks is that they get lots of new options, which vest annually so long as they remain.
Once the deal is signed, both sides often try their best to wiggle out. The stock options aren't paid out if the employee quits early, so the company tries to get the employee to quit. CEOs become directors of empty divisions with no staff and no mission, stuff like that. The company can't be _too_ blantant about it (i.e. make the CEO unblock toilets all day) as that's constructive dismissal, in which case the employee can leave with the stock (after lots of legal squabbling, of course). Equally, mr small-company-entrepreneur type wants to get the stock and bug out (either to his next startup or to Hawaii) and doesn't want to be a drone for the next half decade. So he _tries_ to get constructively dismissed. Fired for gross misconduct (not showing up, punching out his boss, etc.) won't work - so he has a bad attitude, doesn't bathe, says dumb things to the media, produces product that makes his employer uncomfortable, founds the aryian-spaceship-league, whatever. So a war of attrition is fought.
Naturally, I don't know the terms of the nullsoft acquisition, but it may be this is Frankel's (et al) idea (or at least in his mind). I figured this was the case when Gnutella came out (AOL were _never_ going to be happy with that) and WASTE is even more AOL-unfriendly (heck, it's got a chat client - who needs AIM?).
Someone should write a book about the constructive dismissal stories that fill Silicon Valley - Sculley sending Jobs to his own office building to do nothing (Jobs cracked rather quickly). I heard of some guy coming to work dressed in a full frogman suit (including flippers and mask) and walking down in the corridor when customers were around - company dress code said "no shorts, wear shoes" - if they'd changed it to read "no bodyglove swimming attire" just for him, then that would have been the constructive dismissal he sought.
If Sony were the monolith its detractors claim, the PS developers would be pretty much mandated to run ITRON/uITRON/JTRON in this kind of device. I think Sony licences an emotion engline uITRON build to games developers who want one - does anyone know for sure what RTOSes do work on EE?
For regular computer display purposes, a transparent screen doesn't seem terribly useful, due to contrast and "visual noise" interference from whatever is behind the screen (mitigated a bit if the screen is frosted).
Still, there's plenty of possible applications for this:
I can't believe no-one's said it already, so looks like it's down to me to trollishly quote The Simpsons - "He's gradually getting away!"
There's a lot of work being done on reconfigurable computing, which imagines replacing the CPU, GPU, DSP, soundcard, etc., with a single reconfigurable gate array (like an RAM-FPGA). You'd probably have a small control processor that manages the main array. On this array one could build a CPU (or several) of whatever ISA you needed, and GPU, DSP, whatever functionality was called for by the program(s) you're running at the current moment. Shutdown UnrealTournament 2009 and open Mathlab, and DynamicLinux will wipe out its shader code and vector pipelines, and grow a bunch of FP units instead. Run MAME and it will install appropriate CPUs and other hardware.
In the initial case, this would be controlled statically, a bit like the way a current OS's VM manages physical and virtual memory. Later, specialist "hardware" could be created, compiled, and optimised, based on an examination of how the program actually runs (a bit like a java dynamic compiler). So rather than running SETI-at-home your system would have built a specialist seti-ASIC on its main array. There will be lots of applications where most of the work is done in such a soft ASIC, and only a small proportion is done on a (commensuately puny) soft-CPU.
This all sounds too cool to be true, and at the moment it is. Existing programmable gate hardware is very expensive, of limited size (maybe enough to hold a 386?), runs crazy hot, and doesn't run nearly quickly enough.
It adds another weird element to the already surreal aspect that Ames/Moffet presents, particularly to the north. There's a number of odd (nay, sinister) looking buildings, some positively Quatermassey domes, weird towers, and of course the giant rectangular intake of the wind-tunnel building. The whole place has a cool area 51 big science of the 60s feel about it.
Combine that with the Mountain View city lot beside it, where they keep hundreds of trees and bushes in wooden boxes, ready to be transplanted, lined up in neat little rows - it looks a bit like the set of The Prisoner.
Nearby is SGI's main campus, where they've build a couple of ultra-modern office buildings (not as short of cash as we may have thought). Given that SGI's major remaining customers are NASA and NSA, it's get another little piece of the "look what government money built" zone up by Shoreline.
There are several free, reasonably mature windowing environments available for linux already, many featuring hardware acceleration. Several are suitable for embedded use. Why do I want to spend $40 for this? (I'm not being rhetorical - the site isn't accessable). There are innumerable linux distributions, several of which boot straight from CD without install.
Frankly, the speed differentials Scitech quotes (over Xfree) aren't really all that impressive for most graphics adaptors. Sure, there's a big difference between unaccelerated (e.g. vesa) access and accelerated, but a 20% differential between the 2D performance of one accelerated solution and another just isn't that compelling. Now many applications are _that_ dependant on 2D performance? If I'm that 2D bound I can spend that $40 and get vastly better graphics performance by buying a better card.
The "foo is old fashioned", "foo is too complicated", "foo is SO last century" claims that some make (I dunno if these guys do, as their site is still down) aren't value propositions. Is something significantly faster? Significantly smaller? Significantly more useful features? Significantly cheaper? Those are.
Parenthetically, note that I don't apply this standard to free projects. Someone can go code a new OS just for their own pleasure, and doesn't have to pass a customer-value-proposition test. Why? Cos they don't have customers, and so they're not obligated to provide value to anyone.
While the QNX stuff is generally of excellent quality the FlexLM thing is a persistent source of problems. Installation and upgrade have never gone smoothly, with obscure services not starting or being misconfigured by the installer, client authentication going wrong more times than it should (i.e. ever), and occasional file-locking problems that require a reboot. At least in my case, licence management seems to generate as much traffic with the QNX support folks as does their actual product (host and target) in its entirety.
Worryingly, the licence is bound specifically to one licence server. I _imagine_ that if the machine (a laptop) were destroyed, lost, or updated, then there would be some means whereby I could persuade them to issue another licence, but it's bound to be a sticky point.
I wouldn't care if everything worked properly and transparently, but it doesn't. My vendor is essentially treating me like a thief and simultaneously making himself look like a bozo (which he isn't - the rest of the QNX stuff is great).
On the last occasion it took several days to resolve the licence manager issues - had this been at a more critical time then this would have been a dealbreaker. It leaves me with a rather bitter taste in my mouth, and I'd think twice before recommending QNX to another client, purely for this reason.
So is their bizantine DRM saving them money, or costing them? I think Intuit can answer that for them.
Either to engage in, or to defend, litigation a large corporation would generally engage a law firm specialising in litigation, and probably one specialising in the particular type of litigation - employment, competition, IP, environmental, etc.
So the question should be "how big is apple's legal budget?".
The Sun java folks used to (until last year) work from two buildings on DeAnza in Cupertino. At least one of these was an ex-Taligent building, and consequently IBM owned the furniture (I think the lease may also have been some kind of sublease thing). One time IBM wanted all their furniture back, and I believe they flat refused to sell it, forcing each Sun java employee to move out of his office into the corridor, while the facilities dudes came and swapped his desk etc. out for an essentially identical replacement.
The sun java folks are now confined largely to Sun's Agnew's development centre, built on the site of the county mental hospital. Given that Cupertino was a totally excellent place to work, and Santa Clara most assuredly isn't, I'd guess that if the IBM folks said "we'll buy java, and y'all can come back and work in Cupertino" there would be a lot of happy people.
One of IBM's largest Java development centres is (waitforit) on DeAnza in Cupertino, right beside the old Sun java building. Both are former Apple buildings, and a bunch of the java folks are ex-apple.
I wouldn't put too much stead in the "disgruntled employees veto the deal" theory, mentioned above. These days, the average Silicon Valley employee cares about 1) do I get paid ? 2) does my commute get better or worse ? 3) do I get to do something that isn't totally crap ? (the former number 1, "will my stock options make me rich?" no longer figures much).
Rickover took the seal guys aside, and asked them - if your son was on this boat, would you still want seals, or would you opt for the magnetic method? The seal guys thought for a while, and sheepishly replied that they'd go with the magnets. To this day, all US naval reactors have magnetic interlocks, not seals.
Fact is - seals are hard. Hard to make, hard to maintain, and hard to check. They're almost always the first thing to fail, and rarely gracefully.
So, rather than the next gen spaceplane being some slicko streamline hitech composite fibre whatnot, it should be a windowless monocoque made from thick polymerised concrete. The astronauts will need a stihlsaw to go EVA, but then a concrete spaceship needs no maintainance, so they won't have to.
Good forks have the following in common:
Consider some good forks:
- mozilla -> phoenix -> mozilla(whateveritisbird)
- X11 -> XFree
- GCC2 -> EGCS -> GCC3
- linux is perhaps the best example - two major branches running all the time, and both
(particularly the 2.3, 2.5, etc. dev fork) heavily forked themselves. 2.5 changes are
often backported to 2.4, even to 2.2, and the maintainers all still talk to one another.
By way of contrast, the GNUemacs/Xemacs fork is a prime example of a bad fork. Bad blood, wilful incompatibility, divergence, duplication of effort.If XFree's current "governance fork" turns into an all out code fork then that would, I fear, be a bad fork - all that bad blood will surely make things very difficult technically.
So perhaps the best advice to a successful project is "encourage forks, and provide a safe environment for them". Apache and Mozilla both do this, to their benefit and credit.
On a related note, can anyone recommend a decent open source / free software graphical font design tool ? I looked into this a few years ago and things deemed to be in a crufty state of disarray. What do folks use now?
Does anyone remember this (or is it just my imagination), and if so, what became of it?