I've got a beloved Cobalt RaQ4 running a proprietary app server.
I've booted it once when I turned it on an that's it. That was a year and a half ago. I patch it when necessary, keep the fluff out of it's inlets and that's it. (Sometimes I stroke it, and sing to it)
On the other hand I have the same app server running on a Windows development box and, well, you can just tell what I'm going to say so I won't bother.
I saw some of these a CEBIT last year and the projector is a few inches above the desk and angled down at about 45deg. Didn't get to try any though.
You've got to interrupt the beam of several letters just to "press" one key, so it must be pretty "smart" just to work that out.
I guess it can assume it's on a flat surface but if it doesn't know how fat your fingers are then it doesn't know when you've touched the table either.
Maybe there's a secondary scanner at table height that returns the x/y of anything that interrupts the beam?
Even if a system could be invented that guarantees security, integrity and privacy, the "proof" that it could be trusted would be beyond the man in the street.
Everyone (well nearly everyone) can see and understand Xs, bit's of paper, security vans and vote counting.
Try explaining non-repudiation, PKI infrastructure and certification to one of your maiden aunts.
Will she be more or less convinced that the next President really won?
If people don't understand it they won't trust it. And if they don't trust it they won't use it.
The 15-year-old-girl in South Africa who suffered second-degree burns to her hands and thighs after her laptop exploded was using a Dell machine, it's been revealed.
iafrica.com reports that the laptop was a Dell Inspiron.
The story also quotes the girl's father as saying: "It would appear the battery pack had exploded and disintegrated and set fire to the room."
Forensic experts are now carrying out tests on the laptop to establish the cause of the explosion and subsequent fire.
A spokesman for Dell in South Africa said the company had not been contacted by the family or police over the incident.
And because the full facts of the case had yet to appear he was "unable to comment" on the matter.
1. Just ship suns VM and break apps that rely on the MS VM.
2. Ship the MS VM AND the sun VM and allow the end user to choose which is active in the browser (like the Java PlugIn does).
3. Add the MS "extensions" to the Sun VM so MS "Java" apps/applets work with that or can at least be easily ported.
Options 2 and 3 are obviously best for their customers, but both imply continued investment in Java by Microsoft, which I'm sure pleases them immensely (that was irony by the way).
> If you're going to write a Java program, then you should write it to run in Java. Not MS' "Java"
I agree, assuming you know what you're doing - but not everyone does.
Some new graduates working on a project in a company I worked for were (it transpires) using Visual J++ (why!) and had "no idea" it wouldn't work anywhere else until they shipped it.
It doesn't say much for their project manager I know, but these issues tend not to get discussed at university because they're "political".
As has been said, GSM is (shock horror) quite a good voice protocol.
It can also hand over between cells.
It can also support micro-cells inside buildings etc to fix congestion problems.
Surely enough money has been wasted in the telecoms sector. Lets wait until some truly revolutionary concepts are developed, something that brings genuinely new and valuable features to the market place before rolling out another half baked infrastructure.
An aspect of programming that doesn't get discussed much is how they interact with out mental frailties.
Arguments about the best/fastest/smallest/cleanest language generally play to our personal predjudices.
The need for self expression often motivates us to choose a language, maybe you like objects or regular expressions. Maybe compilers are a drag or you never quite fathomed all that multiple inheritance syntax.
But overall we're flawed as a species and tend to make the same mistakes over and over:
We are forgetful: hence C prototypes show us when we've forgotten what function parameters should be.
We are careless: hence Java exceptions force us to think about (if not do anything about) error handling.
We need structure to aid comprehension: hence while/repeat/for etc when goto and if work just fine.
etc. etc.
In my view a good language is one that takes account of our stupidity and compensates for it, avoids it or slaps us in the face before we ship it to customers.
Frankly, languages like C and particularly C++ are poor in this regard when compared to languages like Java.
Sure, a Godlike coder will produce a smaller, faster solution in assembler, C or C++. But what about the rest of us that forget to free strings in our destructors, accidentally introduce side effects into out macros or have creative mishaps with CVS?
I remember reading (about a decade ago now on comp.risks I beleive) the following:
BMW released their new micro-processor (ooooh!) controlled ABS braking system when odd transient failures were reported by some customers - those that survived anyway.
It turns out that in certain excessively high RF environments the processor locked up. One such environment was while driving past one of those huge "Golf Ball" early warning radar stations so "popular" at the time.
I remember this because of the interesting discussion of what the fail-safe behaviour of a braking system ought to be:
void brakesPressed(BrakeEvent e)
{
try
{
setBrakesLevel (e.getFootPressure ());
}
catch (GoneHorriblyWrongException oops)
{
// setBrakesLevel (0.0);// killed the test team
// setBrakesLevel (1.0);// got rear ended by a truck
setBrakesLevel (0.3);// lets see what happens...
> What does the/. community think of the growing move toward the web platform.
I'm not sure what a true "web platform" really is. If you were to design the thinnest possible platform that supported "reasonable" quality browsing and not much more, what would it be, and how small?
So far the best attempts I've seen have been built on heavily pruned Linux, just enough to run Mozilla or Opera. But that's still fairly fat, a side effect of being a modular OS that's capable of so much more.
I'm still waiting for a descent "web platform" that fits on a floppy or two, I'm confident it could be done though, given a fixed hardware platform.
No doubt there are several/. ers out there who wrote one just last week:-)
Without wishing to diss HP (I worked for them for several years and sadly I'm still a shareholder) I wouldn't be quite so confident.
HP has a long an illustrious history of doing the following:
a) Developing genuinely great technology. b) Sitting on it for too long because the division heads don't believe in it. c) Inventing a way to derive a revenue stream from the product that makes it look bad in the market place. d) Finally releasing it in a butt ugly box. e) Canning it after 6 months cos it didn't make a billion.
An interesting contrast to the way Sony does things I think, look at the sustained commitment they have to their technologies!
For the sake of my shares I hope I'm wrong this time.
> Hm, can't find a friggin flashlight when I need one. Guess I'd better print one out... > Can't afford the one from maglite.com, cool as it is... What to do? > Ah! Of course! Download the open source flashlight from opengadgets.com and print it out.
An interesting copyright issue:
Are the *PLANS* in breach of copyright if the *PRODUCT* would be *IF* printed.
Suppose the design file (it's DNA?) would literally produce a MagLite, identical to the genuine article, yet were not derived from the MagLite company. Would the plans themselves be "legal"?
Another analogy: I might have been named George Bush by my mother. I might naturally resemble him and have an overwhelming desire to consume pretzels without using my teeth. Should my parents be arrested?
> Just add a GPS, and your employer will know you're at the bar instead of the conference room.
Once stayed in a Hotel, "back in the analogue days" and was having trouble with the TV reception. Pulled out the TV and found the screws at the back that tuned the posts for the channels.
Soon discovered that I was getting interference with the set top box from the room next door, after all it was probably only a couple of feet away and the cables were low quality.
A little fiddling and pretty soon I could tune in to S.T.B. next door with "reasonable" quality and see exactly what he was watching. Back To The Future III over and over and the first free minutes of the pron channels.
Can't think what made me think of this in a discussion of hotel WIFI...
Lets start an "Portable Buffer Management Library" project on SourceForge, pick a license that's acceptible to everyone and fix this problem once and for all.
How hard can it be?
Lets start now and take it in turns, this is my contribution:
#include
Aw damn I've alienated Visual C++ coders already...
Slashdot Mirror
... will you see what you're typing on your projected keyboard?
> Huh? I guess the only Windows system you have ever maintained is your mom's WindowsME.
I remember one delicious incident supporting a windows server.
1. Needed to upgrade the database.
2. Forced to install IE to do so.
3. IE won't run because Graphics card not good enough.
> What about the GUI?
What GUI? It doesn't have a monitor...
I've got a beloved Cobalt RaQ4 running a proprietary app server.
I've booted it once when I turned it on an that's it. That was a year and a half ago. I patch it when necessary, keep the fluff out of it's inlets and that's it. (Sometimes I stroke it, and sing to it)
On the other hand I have the same app server running on a Windows development box and, well, you can just tell what I'm going to say so I won't bother.
I've avoided several otherwise promising server products because they only run on Windows.
I like to spend my weekends with my family, not hunting for boot disks at 3am on Saturday morning in an darkened office block.
I hope the trend continues...
I can't think of a scheme that seems reliable.
I saw some of these a CEBIT last year and the projector is a few inches above the desk and angled down at about 45deg. Didn't get to try any though.
You've got to interrupt the beam of several letters just to "press" one key, so it must be pretty "smart" just to work that out.
I guess it can assume it's on a flat surface but if it doesn't know how fat your fingers are then it doesn't know when you've touched the table either.
Maybe there's a secondary scanner at table height that returns the x/y of anything that interrupts the beam?
I want to see how the stabilizing circuitry of one of these babies handles a pot shot from my home made EMP rifle.
When's open season?
> Internet voting, along with electronic voting booths at supermarkets was used in St. Albans, U.K. last year.
Where Labour won by a majority of 76%, the Conservative party coming a close second with 72%.
Even if a system could be invented that guarantees security, integrity and privacy, the "proof" that it could be trusted would be beyond the man in the street.
Everyone (well nearly everyone) can see and understand Xs, bit's of paper, security vans and vote counting.
Try explaining non-repudiation, PKI infrastructure and certification to one of your maiden aunts.
Will she be more or less convinced that the next President really won?
If people don't understand it they won't trust it. And if they don't trust it they won't use it.
VoterApathy*=2;
All of the "saucers" were seen from the side and nearly everything was symmetrical and about the X and or Y axis.
If these things were just zipping about we'd see them from all sorts of angles?
Surely this implies some artifact of the CCD properties or post processing?
And, unless these things are all the size of pluto, they'd almost be invisible to the camera or only a pixel big anyway.
On a sombre note, from a register article:
The 15-year-old-girl in South Africa who suffered second-degree burns to her hands and thighs after her laptop exploded was using a Dell machine, it's been revealed.
iafrica.com reports that the laptop was a Dell Inspiron.
The story also quotes the girl's father as saying: "It would appear the battery pack had exploded and disintegrated and set fire to the room."
Forensic experts are now carrying out tests on the laptop to establish the cause of the explosion and subsequent fire.
A spokesman for Dell in South Africa said the company had not been contacted by the family or police over the incident.
And because the full facts of the case had yet to appear he was "unable to comment" on the matter.
It has to be made law since it benefits society as a whole at the expense of the individual (hey - I'm European!)
For example: I'm pushing for a new car safety law - I know it initially sounds counter intuitive, but bear with me.
Firstly seatbelts, front crumple zones and air bags are made illegal.
Secondly the steering wheels of all cars are replaced by ones that have a single 3 inch long rusty spike in the middle, facing the drivers chest.
Obviously nobody would *CHOOSE* to do this but the world would clearly become a better place if it were a legal obligation.
I was once on an interview panel (does that make me responsible?) when the interviewees mobile phone wrang.
... great ... ... look, I'd better better call you back later, I'm a bit busy right now".
He answered it.
Then gestured to one of us to give him a pen!
He then proceded to write down a phone number and said something like:
"so then 7.30, is John coming?
He then hung up and said:
"Sorry, what was that last question"?
Did he get the job?
What do you think?
It seems MS have several options here:
1. Just ship suns VM and break apps that rely on the MS VM.
2. Ship the MS VM AND the sun VM and allow the end user to choose which is active in the browser (like the Java PlugIn does).
3. Add the MS "extensions" to the Sun VM so MS "Java" apps/applets work with that or can at least be easily ported.
Options 2 and 3 are obviously best for their customers, but both imply continued investment in Java by Microsoft, which I'm sure pleases them immensely (that was irony by the way).
> If you're going to write a Java program, then you should write it to run in Java. Not MS' "Java"
I agree, assuming you know what you're doing - but not everyone does.
Some new graduates working on a project in a company I worked for were (it transpires) using Visual J++ (why!) and had "no idea" it wouldn't work anywhere else until they shipped it.
It doesn't say much for their project manager I know, but these issues tend not to get discussed at university because they're "political".
As has been said, GSM is (shock horror) quite a good voice protocol.
It can also hand over between cells.
It can also support micro-cells inside buildings etc to fix congestion problems.
Surely enough money has been wasted in the telecoms sector. Lets wait until some truly revolutionary concepts are developed, something that brings genuinely new and valuable features to the market place before rolling out another half baked infrastructure.
And don't talk to me about WAP or MMS!
The only safe web server is an un-installed web server.
An aspect of programming that doesn't get discussed much is how they interact with out mental frailties.
Arguments about the best/fastest/smallest/cleanest language generally play to our personal predjudices.
The need for self expression often motivates us to choose a language, maybe you like objects or regular expressions. Maybe compilers are a drag or you never quite fathomed all that multiple inheritance syntax.
But overall we're flawed as a species and tend to make the same mistakes over and over:
We are forgetful: hence C prototypes show us when we've forgotten what function parameters should be.
We are careless: hence Java exceptions force us to think about (if not do anything about) error handling.
We need structure to aid comprehension: hence while/repeat/for etc when goto and if work just fine.
etc. etc.
In my view a good language is one that takes account of our stupidity and compensates for it, avoids it or slaps us in the face before we ship it to customers.
Frankly, languages like C and particularly C++ are poor in this regard when compared to languages like Java.
Sure, a Godlike coder will produce a smaller, faster solution in assembler, C or C++. But what about the rest of us that forget to free strings in our destructors, accidentally introduce side effects into out macros or have creative mishaps with CVS?
> Who the hell wants a watch that crashes with the BSOD whenever I ask it to tell me the time!
Don't be churlish, the weekly service patches will solve that problem in no time.
What concerns me is it's accuracy as a basic watch, everything else out of Redmond runs slower and slower over time until it's bi-annual reinstall.
BMW released their new micro-processor (ooooh!) controlled ABS braking system when odd transient failures were reported by some customers - those that survived anyway.
It turns out that in certain excessively high RF environments the processor locked up. One such environment was while driving past one of those huge "Golf Ball" early warning radar stations so "popular" at the time.
I remember this because of the interesting discussion of what the fail-safe behaviour of a braking system ought to be:
void brakesPressed(BrakeEvent e)
}{
Considering the first point:
/. community think of the growing move toward the web platform.
/. ers out there who wrote one just last week :-)
> What does the
I'm not sure what a true "web platform" really is. If you were to design the thinnest possible platform that supported "reasonable" quality browsing and not much more, what would it be, and how small?
So far the best attempts I've seen have been built on heavily pruned Linux, just enough to run Mozilla or Opera. But that's still fairly fat, a side effect of being a modular OS that's capable of so much more.
I'm still waiting for a descent "web platform" that fits on a floppy or two, I'm confident it could be done though, given a fixed hardware platform.
No doubt there are several
Without wishing to diss HP (I worked for them for several years and sadly I'm still a shareholder) I wouldn't be quite so confident.
HP has a long an illustrious history of doing the following:
a) Developing genuinely great technology.
b) Sitting on it for too long because the division heads don't believe in it.
c) Inventing a way to derive a revenue stream from the product that makes it look bad in the market place.
d) Finally releasing it in a butt ugly box.
e) Canning it after 6 months cos it didn't make a billion.
An interesting contrast to the way Sony does things I think, look at the sustained commitment they have to their technologies!
For the sake of my shares I hope I'm wrong this time.
Maybe when it powers up for the first time it can print out a cable and driver disks.
And some spare cartridges...
> Hm, can't find a friggin flashlight when I need one. Guess I'd better print one out...
> Can't afford the one from maglite.com, cool as it is... What to do?
> Ah! Of course! Download the open source flashlight from opengadgets.com and print it out.
An interesting copyright issue:
Are the *PLANS* in breach of copyright if the *PRODUCT* would be *IF* printed.
Suppose the design file (it's DNA?) would literally produce a MagLite, identical to the genuine article, yet were not derived from the MagLite company. Would the plans themselves be "legal"?
Another analogy: I might have been named George Bush by my mother. I might naturally resemble him and have an overwhelming desire to consume pretzels without using my teeth. Should my parents be arrested?
> Just add a GPS, and your employer will know you're at the bar instead of the conference room.
Once stayed in a Hotel, "back in the analogue days" and was having trouble with the TV reception. Pulled out the TV and found the screws at the back that tuned the posts for the channels.
Soon discovered that I was getting interference with the set top box from the room next door, after all it was probably only a couple of feet away and the cables were low quality.
A little fiddling and pretty soon I could tune in to S.T.B. next door with "reasonable" quality and see exactly what he was watching. Back To The Future III over and over and the first free minutes of the pron channels.
Can't think what made me think of this in a discussion of hotel WIFI...
Enough!
Lets start an "Portable Buffer Management Library" project on SourceForge, pick a license that's acceptible to everyone and fix this problem once and for all.
How hard can it be?
Lets start now and take it in turns, this is my contribution:
#include
Aw damn I've alienated Visual C++ coders already...