Posts by Charles Manning 3509 publicly visible posts • joined 26 Jun 2007
would be being discovered by an alien archaeologist trying out some ancient radio frequency kit that the aliens stopped using 20,000 years ago because they found better ways to communicate. Then getting gentrified by aliens moving in and taking over our quaint customs and lifestyle.
If you use human history as a parallel for what alien history might be like: we've been intelligent for 10,000 or so years. We've only been making reasonable levels of RF energy for less than 100 years. That's about 1% of the time.
So even if there is an intelligent alien population there is only a very small chance that they are doing detectable RF.
Overhang, where you pre-announce a new product that obsoletes the current product, is a marketing no-no.
But MS normally take it a step further by telling you that you need the new stuff because the old offering is crap.
Here MS take this to the extreme. They overhang the new 6.5 products as they are launched: "Here are all the latest products. They are shit because W7 is coming RSN.". Why would anyone want to buy a new 6.5 if they know it is going to be obsolete and lack the special 7.x features. MS should at least have worked with the handset makers to build a range of transitionary products that you can run 6.5 on but upgrade to 7. Why would any handset vendor want to sign up to any program that treats them this badly?
Average EURO31k/year programmers include Visual Basic developers and other low-skilled folk. Kernel developers are not found in the bargain bin. You probably want to double the programmer fees.
Much of the value in Linux comes from all the testing which never really gets factored in. While programmers' contributions are protected by copyright there is no such recognition or protection for the people that test the code.
If you have chosen the 5000 well, then that is a good enough number to get within very few %. Read http://en.wikipedia.org/wiki/Sample_size
However the problem is choosing well. The sample has to be a random sample of all USA.
For example, if this was a phone survey then that gets skewed by many factors: many people don't have land lines and they're more likely to be high tech users, heavy dial up users were probably busy when called and people with a life probably told the caller to get stuffed. Time of day impacts greatly: probably a high degree of correlation between Luddites that sit watching TV and being home when the telemarketer calls.
Likewise, a sampling on a street corner only gets you the people that pass by.
Most of these surveys are performed by people who really don't care about the quality. They just want to get n calls so they can finish the survey and go home.
1) The number is wrong. While 11,000 lines of patches might be submitted on all mailing lists etc every day, nowhere near as much actually makes it into the kernel. A lot of what does get accepted is actually modifications and refactoring, not new code per se.
2) Lots of it is optional. It includes whole rafts of stuff for usb devices, flash drivers and support for tens of CPU architecture and over a hundred variants of those. There's support for 30-odd file systems (which all exist because they serve some purpose). Only a small amount of the whole code base actually runs on a Linux server, phone or any one device. The kernel code actually running on an Android phone is probably around 750k lines or so.
3) How many CPU architectures did 4.2BSD support? How many USB devices? Audio devices? Flash memory devices? How many different file systems? It's pointless comparing them.
4)The reason there are so many monkeys coding Linux is because there are so many variants and it is being used in all sorts of applications. A relatively small number are actually generating stuff that **everyone** uses.
WM has never gained traction as user-buys consumer phones. They have always been business oriented phones bought by corporates. IIRC, Ballmer even predicted that the over-cute iphone would never be used as a serious phone by serious business types because of its flair.
Adding all the consumer-oriented eye candy might slightly increase appeal to the Xbox brigade, but that's surely going to erode any market share they had in the serious phone space. Business users will surely switch in droves to RIM (if they want standardised email etc) and iphone or Android or Symbian if they want turnkey or specialised applications as well.
This is a really poor attempt to rebrand and relaunch WM into a different market. It scews their current market and is insufficient to make inroads against existing players.
Chalk this up as another cock-up on Ballmer's watch.
What's a credible second place in a running race or any other competitive game? Perhaps performing at 90%+ the speed of the winner.
Performing at around 50% of the winner is hardly second place. It is just head of the loser pack.
MS+Yahoo are now only 60-odd percent of what they could have been two years back so they're heading in the wrong direction. Together they'll sink faster.
As you rightly say, the heat dissipates into the whole server pretty much the same whether it comes from a point source or a raft of heat sinks. The temperature of the whole server is a function of the power of all the boards and internal heat sinks only protect individual chips.
About the only real issue is in the area very close to a high temperature part. High temperatures can cause mechanical stresses in the solder joints, pcb vias etc. If there is difference in failure patterns then that's the most likely cause.
And size, cost and heat management.
While Atoms might be low power compared with desktop x86 chips, compared with ARM they are terrible. Atoms boast 4W or so but still need a 6W or so companion chip + all the associated circuitry to make them work. That makes big/heavy/expensive boards, big/heavy/expensive batteries and big/heavy/expensive mechanicals
The equivalent ARM-based solution is sub-2W and small. That means you don't need as big a power supply section and less thermal management (ie. smaller/lighter/cheaper circuit boards and heat sinks). And a smaller battery will suffice (smaller/lighter/cheaper again). That in turn means smaller/lighter/cheaper mechanicals.
That's the way electronics is: the payback for reducing size, cost and power consumption is reaped multiple times, not just once.
Typing is a Girls Blouse interface and I blame punch cards for starting the rot. Real men don't need anything more than LEDs and toggle switches. Monochrome 1 pixel porn sucks though.
Your point is valid sir, nothing really wrong with flash since it is the best we've had so far as a mechanism to deliver certain content. Of course, like anything, flash can be used in good ways and bad ways and HTML5 or whatever is next will be deployed just as poorly, if not even worse.
Likely much of the stuff written for HTML5 will be ported from flash designs giving you the worst of both worlds.
Funding space science is just funding a bunch of people who have an interest in something that really has little benefit to anyone else. Like funding train spotters, just more expensive.
Space "research" is very expensive and really generates very little useful knowledge. There are far more important areas of research here on earth that are far more relevant and have better payoff.
So Mars has some craters and CO2 and shit... So who really cares? A handful of scientists get a stiffy and work up PhDs but what real benefit is there? It's just infotainment.
Sure, 1960s space race generated an interest in science but most kids really just wanted the cool factor of dressing up in all mom's tin foil and being an astronaut. Now they all want to be rappers or whatever is cool these days.Tthe space race was really a political race. These are once-off events lose their shine pretty quickly. and, like telling a joke, it only really works once. By Apollo 13 they could not even get TV air-time until things broke and the voyeur effect kicked in.
Yeah, we got satellites, which are very handy, but launching satellites was achieved in 1950s and is a completely different thing from sending people to ISS or missions to moon or mars.
Yes, we got some spin-off developments in "space age" materials. But that's not enough. We will get spin-off developments from **any** research effort.
MS has attempted tablets many times, starting way before 2000, and they've all been complete market failures with only a few niche markets finding a use for them. Why? coz just trying to mak a flat PC and calling it a tablet does not work.
MS tried to push the idea that every utility worker and nurse would carry around a tablet instead of notebooks and files. Of course that doesn't work unless the organisations commit to complete computerization efforts and the tablet can go a whole shift on one battery.
You don't want this as a phone or an ipod replacement.... it does not fit in your pocket and it just lacks the keyboard and sheer horsepower to replace a laptop. It will replace none of these and is not intended to [MS's big fail was trying to replace laptops with tablets].
The reality is that a tablet is ergonomically suited to only a certain set of functions and most of those revolve around the novelty gadget functions: the webpad idea that was thrown around in the early 2000s. Basically a big ipod touch that you leave in the lounge to look things up on the www. Maybe also as an ebook reader and a personal video player / limited gaming device at a push. Limited email and twitter etc is probably OK too.
It looks like Apple have taken a pretty good stab at a tablet. What is unclear is whether this is enough to excite people.
Duh!
It has always been about the software, never the hardware [*].
Nobody carries a 2 pound lump, cellphone or even a laptop, around just for the fun of doing so. Nope, they carry it around because it provides life support for the software/data and the function those provide.
Nobody want's the hardware. They just want something that gives them the software. The hardware is a penalty you have to pay to have the software around.
Apple have taken the right route with this. A tablet has far more in common with a phone than a PC. MS have screwed up at least 4 attempts to get into tablets (one of Bill Gate's pet projects) because they have tried to make it into a Windows machine (ie. yet another PC). Of course they tried that on phones too.... Nobody wants to use the start bar on a phone or tablet.
[*] Slight exception maybe for iconic devices, like maybe the iphone, that apparently make you look cool.
Well that explains why it has taken so long. It has taken a while for Moore's Law to make the required resources small enough to fit in the box.
Battery life surely sucks, but it will depend on what it is used for.
What are the primary markets?
I've heard of tablets being touted for use in hospitals etc (every nurse carries one to instantly access and update patient records). But for that you really want something that can last a whole shift (12 hours) on a single battery - two at an absolute stretch.
Umm... surely having a product that forced them to get their shit together is actually a Good Thing?
Note to Reg: When you pick up a fresh batch of hacks at the pub, please let them dry out a bit before they submit articles.
AT&T had an opportunity to build a real cracker business, but would rather not... it just seems too much like effort.
They will undoubtedly lose market share to providers who can and want to provide a superior service.
The only benefit for AT&T is that the monogamy probably prevented them from selling Androids and they can now do so. Not that Android sales will make up for lost iphone sales.
Oh well, AT&T can go back to sleep and ignore those pesky customers who actually want to use their networks.
From a business unit perspective it is important to group products according to the customers, not the internal technology etc. That way your sales/product dev staff etc are able to push a consistent message to customers and internally.
How many Xbox customers also buy Windows Mobile? Grouping them together is stupid.
Windows Mobile is only important in to corporate customers and it thus makes sense to put it in a business unit oriented to corporate customers. That of course does not mean that WM will be re-kerneled with Vista. While smartphones have increased in performance and memory they do no't make quadcore phones yet!
Of course WM !-= WinCE. WinCE might be better placed elsewhere,
How much of a search centre do you get for $1bn? It makes no sense to be a bit-player in search. You're either number one or nothing. As a search user it makes no sense to have to hit 3 small search engines to get all the results when you can just hit Google.
Sure, if Apple had a search centre they'd be able to do searches for the locked in iphone fans, but how long before they start screaming?
The centre is more likely for itunes which generates healthy profits.
How can anyone construe a body scanner to be discriminating? It might be invading privacy, but what has that got to do with discrimination?
Sure a transgender person might prefer that the world does not see his/her penis, and a paraplegic might like to keep his plumbing arrangements hidden. Straight blokes like me might like to cover up man-boobs etc.
If you're transgender, in a wheelchair, white, black, gay, straight whatever the machine works the same. If everyone has their privacy invaded.
Where's the discrimination?
El AL use their own security. Unlike other airlines they don't rely on the security provided by the airport. You'll get a uniform El Al experience regardless of where you board. If El Al can't get the security level they want (eg. due to legal restrictions) they won't fly there.
It's getting to be like USA... Scientists competing for funding sex up their applications by trying to make them current etc. Scientists spending time doing talks and building up their image/reputation rather than getting on with the science.
No wonder the practice of science is far from the ideal of dispassionate search for the truth. Instead it's secrecy, back stabbing and character assassination that looks more like politics or a gangland turf war.
If this is what modern science is, well bugger it. No point in publicly funding it. May as well privatise science and leave it to the corporates.
Most embedded systems don't have any OS at all. Most embedded systems are far too memory/resource constrained to have any OS at all. A typical modern kitchen probably has embedded systems in large and small appliances: stove, fridge, dishwasher, microwave, bread machine, rice cooker, blender,..... none of those is running Linux, WinCE etc. They're mostly running small 8-bit micros with 1kb or less of RAM and a few 4 of flash.
Even your desktop PC has significant numbers of embedded systems: monitor, mouse, power supply, disk drives,...
While your fancy car might have a fancy Linux/WinCE screen in the dash, it is loaded with small embedded systems. Each door probably has its own 8-bit micro to handle the switches/electric window etc. Another few in the dash/console, airbag, lights, seatbelt tensioners,.... These micros are cheaper than wiring.
There are quite a few reasons for this:
* Cost: The cheap micros cost less than 50c US. Adding resources needed to run an OS just add cost with no benefit. Any Moores Law that brings the cost of resources down will also reduce the cost of the small micros to 20c, 10c,...
* Boot time: Bare metal systems can boot up in milliseconds. Linux et al need seconds. That's just not acceptable for any critical function.
* Response time: You want that airbag to fire **now**.
* Simplicity: Simple things are typically more robust. It is a lot easier to verify all the code in a system with only 200 lines. Not so if you're running the whole of Linux etc.
So why does the Reg kick up such a poor survey? Well very few of the people doing real embedded work will be reading El Reg. I would not expect a useful output from Reg readers on embedded systems any more than I'd expect good a good result if the survey was run in Knitting Monthly.
When I was growing up, "as busy as a beaver" would mean working hard. I stopped using that phrase when a Californian lass got all steamed when I said she'd been working like a beaver.
I latter took her aside to explain to her why she should stop telling people that she had a camera in her fanny pack.
Fuck the customers for long enough and they will switch when they get an opportunity.
People trust Google more than they trust telcos. The Telcos have had years of near monopoly to build loyal customer bases but instead have built an industry where poor support and customer screwing are legendary and Google looks like the gentle giant that is going to help the customers. The telcos created the opportunity for anyone and Google have filled it.
Google releases a phone and others will continue to release theirs. This is hardly going to stiff the handset manufacturers any more than Microsoft is screwing the mouse market by releasing Microsoft mice.
Having Al Gore on the Apple board has got to be worth a few Green Points.
After 15-odd years of being a Greenpeace supporter (financial and otherwise) I've stopped. I think they used to do better work than they're doing now. They used to promote all sorts of causes but now seem to have put most of their efforts into promoting global warming which makes them pretty pointless.
Greenpeace's most dopey project has been the Black Pixel project which will supposedly save the world by using less power on CRT screens. The savings are so paltry that almost ny other act of cutting back will be more beneficial. One cup of coffee less per year results in more saving than running the back pixel saver. Having one shower less in your lifetime has a better saving.
Of course they could argue that Black Pixel is a symbolic act to raise awareness. But having a no-coffee or no-shower day once a year would be more symbolic as well as more practical. It would also mean that more people could participate (ie. not just the people with CRTs).
Unbelievable that it takes so many completely numpty posts before getting to one that actually understands the issue.
The problem is in the RTC driver.
Some RTC chips count in binary. Some count in BCD. If you take a BCD representation of 10 and run it through a binary decoding algorithm you'll get 16.
The problem will only strike devices which have BCD RTC hardware but used the stock WinCE RTC driver that expects binary.
Unfortunately most WinCE platform developers will tend to just use the drivers that MS or the BSP vendor sent.
Some will have tested and discovered the problem and will not be running the stock drivers.
They are measuring shareholder returns (ie. increase in share value).
While your general observation is correct that it is difficult to sustain double-digit growth, it is worth noting that:
1) MS has shown no real signs of life since Ballmer got in the Big Chair. In fact, the share value has decreased.
2) The knee-point in the MS performance graph is right where he joined.
Have a look at MSFT vs NASDAQ on http://www.google.com/finance?client=ob&q=NASDAQ:MSFT
What do you see? Since Ballmer came on board the company has been dead. Bobbing up and down with the Nasdaq tide. Now, for sure, MS represents a percentage of NASDAQ, but you'd still expect to see some reasonable deviations caused by MS activity.... Nothing.
Correct.
Old ladies watching trains go past full/empty were used during WW2 to figure out where troops and ordinance were being assembled.
Knowing which bunkers/guns have been seen by drones helps you know which ones haven't and how to plan deceptive games.
Again it would seem that drones have been cast as some magic technology, but don't do what their PR claims. Same happened for Patriot missiles broken system clock etc.
Using Lejos you can run meaningful bodies of Java code on a Lego NXT (64k RAM, 256k flash - with only a small fraction of the flash being used). Indeed, Lejos would even run on the Lego RCX with 32kB of ram and only a boot flash.
But, I agree with your sentiments. Excessive platform wrapping does kill performance.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
