We don't need better programming languages, we need better programmers.
I consider myself a "good programmer". I still code buffer-overflow bugs, but I just mark them with comments as "Fix this later". I'm good because I know what I'm doing wrong, when I do it. I'm bad, because I still don't have the project timescale to code it all "properly" on any project I've worked on over the last 10 years.
If you want reliable bounds checking, then you have to have it supplied automatically (i.e. Java). There's no time in project schedules for coding it by hand.
If Java puts a 100% overhead on execution times (an excessive estimate), then it's the performance equivalent of shipping your product 18 months later. If your schedule speeds up (because Java codes faster than C++) and the debug time decreases (because Java source is less buggy, and debugs much faster than C++) then you may find that the overall saving approaches a useful fraction of the 18 months, or at least enough of it to have your users live with the slowdown.
C++ can rot in hell (or Redmond, which is plug-compatible), as far as I'm concerned. If I can code it in a loosely typed scripting language like J[ava]script or Python I'll do that, and if I can't then I'll do it in Java.
I also work heavily with XML and RDF. You just don't want to think about doing that in C++ !
My first big piece of Fortran (I'm tired of typing uppercase - get over it) was porting a huge piece of Algol 60, written by generations of physicists, into Fortran because the new "upgraded" ICL machine (an Estriel, when they were still called that) only supported Algol 68, not Algol 60.
Youth of today, they just ask why I didn't recompile it for Algol 68....
#karma_whore If you care enough to read this far, you'll like
this.
They apologise for "giving SPAM a bad name,", which is hardly Unisys' fault, but forget about the only reason anyone knows Unisys' name these days - the.GIF patent.
I smell PR Bunnies at work - although time was when WiReD would have called them on something like that.
We have a slightly nicer front end (pre-compiler) called RATFOR
FMR ! I'm used to hearing the "FORTRAN 90, it's not as horrid as it used to be" viewpoint, but I'm amazed that anyone still used RATFOR. I ported RATFOR (from The Book) to George 3, back in the mid-80's. At the time, it was a huge improvement over the FORTRAN I was using (can't remember what, but RATFOR was originally built for the WatFOR and FORTRAN V generation) but even FORTRAN 77 and 80 seemed to have overtaken it for "structured programming" constructs.
You shouldn't still need RATFOR. Not even FORTRAM shops should still need RATFOR.
It has to do it reliably, and fast.
FORTRAN does that fine. However, programs also need to be affordable to build and maintain - FORTRAN is terrible for this. In the last decade, we've also become more interested in connectivity and interworking between systems. Again, FORTRAN is terrible here (although I did hear from a brave chap building an XML DOM in FORTRAN).
FORTRAN is still useful, but only because of the legacy code. I'm not even sure that's still viable (automatic porting works pretty well, when it's heading upscale).
The real intelligence test in the Mensa application process is realising that you don't need to join a society to show that you are clever
Although joining Mensa is a good way to meet real, live, humans - something for which that is clearly the only route available, for far too many Mensans.
I just left (UKM) because I didn't like the far right politics that was far too prevalent there in the mid-80's.
I worked for A Large Bank in '92. Although the cards supported challenge/response, the guy writing the code for a funds transfer app (several M in a transaction) didn't understand this, and did indeed write it in such a way that sniffing the serial would have allowed spoofing. Equally, he'd stored vital stuff in the wrong part of the card memory, and it was readable too.
We had smartcards to buy lunch in the canteen (a tech trial, more than a sensible way of operating). I blew one of our lunch cards to emulate the funds transfer card, and left it sitting in the test machine. It worked for a week or two until someone noticed it (glad I didn't leave my name on it !)
The non-card version of the same app used a plaintext password over a dialup.
I still don't trust on-line banking. In a decade of contracting, I've worked in too many of their offices to have any faith in their code quality.
My cellphone does work without a service - It's a clock, and I can play Space Invaders on it. My 6 year old's cell phone works without a service too - it lights up, and he can play at making calls with it.
If the limited functionality without a service is all that you need, then why buy the service ?
If a contract offers a particular set of functions without buying the service, then it's a clear breach of that contract to withdraw those service later. AFAIK, the TiVo contract stated that these services might not be available on later boxes, but that they were available on boxes of that generation.
Your "stupidity explains most perceived malice" comment is probably true here, but that doesn't excuse TiVo's behaviour, nor does it remove the loss suffered by existing TiVo owners.
What's the financial status of TiVo ? I'm not implying anything by this, other than caution, but this sort of low-rent money-grabbing trick has been the action of last-resort by an awful lot of cash-strapped dot-coms lately.
For Xmas, a friend gave me a small keyring torch (flashlight) - an extremely bright white LED torch.
I was delighted with this. I (probably my friend too) still remember saving up (a considerable sum!) to buy my very first red LED (a TIL209, of course), when I was a kid. I know just why it's difficult to make white LEDs, either way, and I'm still amazed even by blue ones. Like most children, I always wanted a torch for Xmas. I had torches - they didn't work very well. Dull orange glow, and a short life from their leaking batteries. As a very special present (I'd been in hospital) my parents once gave me a keyring torch just like my new one - a feeble glimmer that lasted no time at all, but I thought it was wonderful.
Now my son (6 years old) plays with this LED torch too. To him it's just a torch, and why shouldn't a torch be tiny, pure bright white light, and have seemingly inexhaustible batteries ?
Coal waste isn't a problem in anything like the same way. Sure, CO2 emissions are a problem, but that's a political agenda I can hear you grinding at from here.
If you find an ancient cache of Pu, in a quantity small enough to pick up in one hand, it may well kill you. Coal spoil does none of this. That's why it's important to mark Pu as hazardous in a long-term manner, but coal can look after itself.
Sure, coal waste can kill - remember Aberfan - but that takes 1,000 ton quantities that any can recognise and avoid. For plutonium, we don't even know reliably at what temperature it will ignite.
I share a somewhat similar interest - Cold War bunkers, old railway tunnels and subterranea. The main UK group, Subterranea Brittanica, has many archives of interesting locations, but suffers this same problem of how much to disclose.
This dispute seems to encapsulate a lot of what's wrong with America these days. 8-(
Why does anyone need an "Official Geocaching Site" ? Get off your fat SUV-encased butts, get out there and be your own "official" leagues and teams. You don't need some corporate Disney-wannabee telling you how to enjoy yourselves!
Geocaching needs a minimum of two people, some cheap tech, and a flyposted wall poster to communicate between them (oh, and several billions of technology funded by those nice people at the military-industrial complex). You don't need an "official" site, a hierarchy, a league, or a figurehead chairman (especially not a self-appointed one).
Ignore geocaching.com. Don't boycott it, that's itself too organised, just go and do something else instead. There's a whole internet to play with (thanks again to those helpful mil-ind people) - read PhilG'sbook, and build your own geocaching list server.
What is it with America, "Land of the Free", that can't even fix itself lunch these days without a degree of regimentation and standardised prole-feeding-centres that would put North Korea to shame ? Did you throw off the yoke of colonial British Redcoats, just so that you could be fed by uniformed redshirts ?
Secondly, the map site is legally screwed. He's not providing map references, he's providing direct references to someone else's collection of information. As any amount of legal precedent has shown, a collected work like this is material protected by copyright (and rightly so).
If this map site just listed links to locations, links as DCMI points or to the Getty Thesaurus, then there would be no problem -- but that's not what it's doing.
The smallest alphabet is that of Solresol, with 7. The "letters" (or segmental phonemes, if you're being picky) in Solresol may be represented in several ways, not just written, and it's fundamental to the language that they're all identical. It's often called a "musical language", because of their ancestry from the Western chromatic scale, but they have equally valid written and spoken forms, even to the tone deaf.
Solresol is interesting for several reasons, although I'd not claim that it has a particular significant future.
First wholly invented language to achieve any sort of widespread acceptance.
First "interlangua", the notion of a translation intermediary language capable of expressing all other language translations
x->y as the sequence of
x->solresol and
solresol->y.
First language to formally separate semantics and encoding, i.e. the musical phoneme is exactly equivalent to the written phoneme (or that phoneme expressed in arranged pebbles, or smell-o-vision). As a result, it's entirely phonetic, but the distinction goes a long way beyond that.
Most ordinary people won't see many restrictions from the current standard
"Most ordinary people" think that a TV-soap is top-quality drama. This isn't about what's "most popular" or even "occasionally used", it's about claiming to offer a complete support that even the academics wil be happy with. So what if it's an obscure issue of interest only to linguists ? - linguists also have needs that are worth supporting, not just sports broadcasters and populist entertainment.
As with anything that pits mainland China against Taiwan, an awful lot of hot-air is going to be generated by the governments. It doesn't change there being a very real, and worthwhile, issue behind all this.
I'm only surprised that North Korea hasn't laid into the debate too. Maybe 50 years just hasn't been enough to produce a convincing, "decadent capitalist Korean dialect of the Southern region".
So what if you've been using Unicode for ages, Unicode can't handle Chinese in a way that can simultaneously satisfy mainland and non-mainland Chinese.
#karma_whore M$oft is what most people use. Doesn't make it right though.
Well obviously he meant "smelt", which was a very subtle reference to the title.
If you haven't already read it, it's a paternity case for Dirk Gently. Although reared by salmon, the eponynmous fish of the title long harboured the suspicion that he was really a herring.
I literally read the "Trilogy" from begining to end
Read ? You mean there's a _book_ of this stuff too ? I thought it was just on the radio !
(and at Theatr Clywd - anyone else remeber the Vogon Poetry Appreciation Chairs, made from forklifts)
Adams was infamous for his writing style (much like my coding style). Fiddle with the damn thing in a perfectionist manner for years, and never actually ship it until your agent pulled it out with pliers. I've not read this ms., but I bet a half-finished Adams' script is already in a far better state than most writer's final drafts.
Besides which, the Adams otaku will never shut up until it's published, spun-off, and there's some crappy plastic action figures to go with it.
It's impossible to crack this stuff. It's so impossible, my cow-orkers build a dozen of these systems a week, and they're absolutely rock-solid bulletproof. We already have the technology to do this, and it's even affordable to put them into consumers pockets. If you want them anonymous, we can do that. If you want them traceable, we can do that. If you want (and this is the clever one) them to be anonymous, yet irrefutable, we can even do that.
Of course you're right - they're easily crackable. Some pointy-hair will misunderstand it, or an overworked contractor in a bank won't be allowed time to do it right, or (FFS!) someone will let M$oft implement any of it. It will be broken, and it will be some stupid trivial hole that does it.
Microsoft may not be very responsive to public opinion, but the British Government sure is.
HMG is pretty responsive to "public opinion", but it's the opinion that the editor of the Sun tells them to have.
Issues ? What are they ? William Hague's new hairdo - now that's an "issue" that's getting the haeadlines today.
Democratic governments have to provide equal access to government services
Why ? Joe Sixpack just doesn't care enough. Our election is going to be fought on the basis (as ever) of tax handouts, and if we're lucky, who's to blame for Foot and Mouth (a bunch of subsidy-scavenging sheep farmers in Cumbria, doing dodgy midnight deals with black sheep).
they risk having a large group of pissed off voters on election day.
How about pissed off IT geeks ? Three times in the last year this government has put me out of business. First IR35 made freelance IT contractors untenable (although it's still OK to work in just the same way if you're a lawyer or an accountant). Then RIP offered me a two year prison sentence for most interesting uses of one-shot crypto (I was just trying to sell music on-line, with a per-play charging model). Now they've lumped all IT "security" people in with nightclub bouncers and parking wheelclampers, so that I can't even login as root unless I've bought a licence and had a police background check. Again, lawyers and accountants serving the same industry have been exempted.
Blair wants 100% of all government services online by 2005.
Blair is still at the "Every school should have its own Information Superhighway" level of understanding. He's so clueless that yesterday he even let M$oft use him as a PR bunny for an Office XP launch photo-op.
Just as we start to build the first city-block trashing mechasaurs, you want to make them as smart as one of Cthulthu's grandchildren ? Isn't a big dumb dinosaur scarey enough for you ?
Never trust anything with tentacles and no backbone.
Slashdot Mirror
We don't need better programming languages, we need better programmers.
I consider myself a "good programmer". I still code buffer-overflow bugs, but I just mark them with comments as "Fix this later". I'm good because I know what I'm doing wrong, when I do it. I'm bad, because I still don't have the project timescale to code it all "properly" on any project I've worked on over the last 10 years.
If you want reliable bounds checking, then you have to have it supplied automatically (i.e. Java). There's no time in project schedules for coding it by hand.
Think of Moore's Law - 18 month doubling times.
If Java puts a 100% overhead on execution times (an excessive estimate), then it's the performance equivalent of shipping your product 18 months later. If your schedule speeds up (because Java codes faster than C++) and the debug time decreases (because Java source is less buggy, and debugs much faster than C++) then you may find that the overall saving approaches a useful fraction of the 18 months, or at least enough of it to have your users live with the slowdown.
C++ can rot in hell (or Redmond, which is plug-compatible), as far as I'm concerned. If I can code it in a loosely typed scripting language like J[ava]script or Python I'll do that, and if I can't then I'll do it in Java.
I also work heavily with XML and RDF. You just don't want to think about doing that in C++ !
That's a better line if you can spell "Planck"
Oh, and you're probably a bit out on the value, but then you're probably an American and quoted it in slugs/cubit or something
SI for h (not h bar) is 6.626 E-34 J s (for from-memory accuracy).
My first big piece of Fortran (I'm tired of typing uppercase - get over it) was porting a huge piece of Algol 60, written by generations of physicists, into Fortran because the new "upgraded" ICL machine (an Estriel, when they were still called that) only supported Algol 68, not Algol 60.
Youth of today, they just ask why I didn't recompile it for Algol 68....
#karma_whore
If you care enough to read this far, you'll like this.
They apologise for "giving SPAM a bad name,", which is hardly Unisys' fault, but forget about the only reason anyone knows Unisys' name these days - the .GIF patent.
I smell PR Bunnies at work - although time was when WiReD would have called them on something like that.
We have a slightly nicer front end (pre-compiler) called RATFOR
FMR ! I'm used to hearing the "FORTRAN 90, it's not as horrid as it used to be" viewpoint, but I'm amazed that anyone still used RATFOR. I ported RATFOR (from The Book) to George 3, back in the mid-80's. At the time, it was a huge improvement over the FORTRAN I was using (can't remember what, but RATFOR was originally built for the WatFOR and FORTRAN V generation) but even FORTRAN 77 and 80 seemed to have overtaken it for "structured programming" constructs. You shouldn't still need RATFOR. Not even FORTRAM shops should still need RATFOR.
It has to do it reliably, and fast.
FORTRAN does that fine. However, programs also need to be affordable to build and maintain - FORTRAN is terrible for this. In the last decade, we've also become more interested in connectivity and interworking between systems. Again, FORTRAN is terrible here (although I did hear from a brave chap building an XML DOM in FORTRAN).
FORTRAN is still useful, but only because of the legacy code. I'm not even sure that's still viable (automatic porting works pretty well, when it's heading upscale).
The real intelligence test in the Mensa application process is realising that you don't need to join a society to show that you are clever
Although joining Mensa is a good way to meet real, live, humans - something for which that is clearly the only route available, for far too many Mensans.
I just left (UKM) because I didn't like the far right politics that was far too prevalent there in the mid-80's.
I worked for A Large Bank in '92. Although the cards supported challenge/response, the guy writing the code for a funds transfer app (several M in a transaction) didn't understand this, and did indeed write it in such a way that sniffing the serial would have allowed spoofing. Equally, he'd stored vital stuff in the wrong part of the card memory, and it was readable too.
We had smartcards to buy lunch in the canteen (a tech trial, more than a sensible way of operating). I blew one of our lunch cards to emulate the funds transfer card, and left it sitting in the test machine. It worked for a week or two until someone noticed it (glad I didn't leave my name on it !)
The non-card version of the same app used a plaintext password over a dialup.
I still don't trust on-line banking. In a decade of contracting, I've worked in too many of their offices to have any faith in their code quality.
No, it won't (although you raise a good point). Some obvious disabling has taken place too, to remove this possibility.
I can't afford toy cellphones. Here in the UK, it's cheaper to buy a _real_ on and give him my old one !
My cellphone does work without a service - It's a clock, and I can play Space Invaders on it. My 6 year old's cell phone works without a service too - it lights up, and he can play at making calls with it.
If the limited functionality without a service is all that you need, then why buy the service ?
If a contract offers a particular set of functions without buying the service, then it's a clear breach of that contract to withdraw those service later. AFAIK, the TiVo contract stated that these services might not be available on later boxes, but that they were available on boxes of that generation.
Your "stupidity explains most perceived malice" comment is probably true here, but that doesn't excuse TiVo's behaviour, nor does it remove the loss suffered by existing TiVo owners.
What's the financial status of TiVo ? I'm not implying anything by this, other than caution, but this sort of low-rent money-grabbing trick has been the action of last-resort by an awful lot of cash-strapped dot-coms lately.
Smartcards aren't copyable. There's a lot of work in there (especially the JavaCard standards) to make sure they really are secure against copying.
Of course, over-worked contract coders forced to churn code out without time to read the manual fully have a long track record of getting it wrong !
It applies to other technologies too.
For Xmas, a friend gave me a small keyring torch (flashlight) - an extremely bright white LED torch.
I was delighted with this. I (probably my friend too) still remember saving up (a considerable sum!) to buy my very first red LED (a TIL209, of course), when I was a kid. I know just why it's difficult to make white LEDs, either way, and I'm still amazed even by blue ones. Like most children, I always wanted a torch for Xmas. I had torches - they didn't work very well. Dull orange glow, and a short life from their leaking batteries. As a very special present (I'd been in hospital) my parents once gave me a keyring torch just like my new one - a feeble glimmer that lasted no time at all, but I thought it was wonderful.
Now my son (6 years old) plays with this LED torch too. To him it's just a torch, and why shouldn't a torch be tiny, pure bright white light, and have seemingly inexhaustible batteries ?
Coal waste isn't a problem in anything like the same way. Sure, CO2 emissions are a problem, but that's a political agenda I can hear you grinding at from here.
If you find an ancient cache of Pu, in a quantity small enough to pick up in one hand, it may well kill you. Coal spoil does none of this. That's why it's important to mark Pu as hazardous in a long-term manner, but coal can look after itself.
Sure, coal waste can kill - remember Aberfan - but that takes 1,000 ton quantities that any can recognise and avoid. For plutonium, we don't even know reliably at what temperature it will ignite.
Thanks for that
I share a somewhat similar interest - Cold War bunkers, old railway tunnels and subterranea. The main UK group, Subterranea Brittanica, has many archives of interesting locations, but suffers this same problem of how much to disclose.
This dispute seems to encapsulate a lot of what's wrong with America these days. 8-(
Why does anyone need an "Official Geocaching Site" ? Get off your fat SUV-encased butts, get out there and be your own "official" leagues and teams. You don't need some corporate Disney-wannabee telling you how to enjoy yourselves!
Geocaching needs a minimum of two people, some cheap tech, and a flyposted wall poster to communicate between them (oh, and several billions of technology funded by those nice people at the military-industrial complex). You don't need an "official" site, a hierarchy, a league, or a figurehead chairman (especially not a self-appointed one).
Ignore geocaching.com. Don't boycott it, that's itself too organised, just go and do something else instead. There's a whole internet to play with (thanks again to those helpful mil-ind people) - read PhilG's book, and build your own geocaching list server.
What is it with America, "Land of the Free", that can't even fix itself lunch these days without a degree of regimentation and standardised prole-feeding-centres that would put North Korea to shame ? Did you throw off the yoke of colonial British Redcoats, just so that you could be fed by uniformed redshirts ?
Secondly, the map site is legally screwed. He's not providing map references, he's providing direct references to someone else's collection of information. As any amount of legal precedent has shown, a collected work like this is material protected by copyright (and rightly so).
If this map site just listed links to locations, links as DCMI points or to the Getty Thesaurus, then there would be no problem -- but that's not what it's doing.
The smallest alphabet is that of Solresol, with 7. The "letters" (or segmental phonemes, if you're being picky) in Solresol may be represented in several ways, not just written, and it's fundamental to the language that they're all identical. It's often called a "musical language", because of their ancestry from the Western chromatic scale, but they have equally valid written and spoken forms, even to the tone deaf. Solresol is interesting for several reasons, although I'd not claim that it has a particular significant future.
And of course, it's in Unicode too.
On the downside, it's just French with squeaky noises.
Hawaiian is probably the "naturally evolved" human language with the shortest alphabet.
Most ordinary people won't see many restrictions from the current standard
"Most ordinary people" think that a TV-soap is top-quality drama. This isn't about what's "most popular" or even "occasionally used", it's about claiming to offer a complete support that even the academics wil be happy with. So what if it's an obscure issue of interest only to linguists ? - linguists also have needs that are worth supporting, not just sports broadcasters and populist entertainment.
As with anything that pits mainland China against Taiwan, an awful lot of hot-air is going to be generated by the governments. It doesn't change there being a very real, and worthwhile, issue behind all this.
I'm only surprised that North Korea hasn't laid into the debate too. Maybe 50 years just hasn't been enough to produce a convincing, "decadent capitalist Korean dialect of the Southern region".
- You seem to be expecting Unicode to do some magic semantic translation for you.
- Dropping the word "hiragana" into a posting doesn't make you an expert linguist.
- You've entirely ignored the real issue that this article raises.
--"There's something wrong with our bloody moderators today"
Admiral_Jellicoe@slashdot
So what if you've been using Unicode for ages, Unicode can't handle Chinese in a way that can simultaneously satisfy mainland and non-mainland Chinese.
#karma_whore
M$oft is what most people use. Doesn't make it right though.
Well obviously he meant "smelt", which was a very subtle reference to the title.
If you haven't already read it, it's a paternity case for Dirk Gently. Although reared by salmon, the eponynmous fish of the title long harboured the suspicion that he was really a herring.
I literally read the "Trilogy" from begining to end
Read ? You mean there's a _book_ of this stuff too ? I thought it was just on the radio !
(and at Theatr Clywd - anyone else remeber the Vogon Poetry Appreciation Chairs, made from forklifts)
Adams was infamous for his writing style (much like my coding style). Fiddle with the damn thing in a perfectionist manner for years, and never actually ship it until your agent pulled it out with pliers. I've not read this ms., but I bet a half-finished Adams' script is already in a far better state than most writer's final drafts.
Besides which, the Adams otaku will never shut up until it's published, spun-off, and there's some crappy plastic action figures to go with it.
It's impossible to crack this stuff. It's so impossible, my cow-orkers build a dozen of these systems a week, and they're absolutely rock-solid bulletproof. We already have the technology to do this, and it's even affordable to put them into consumers pockets. If you want them anonymous, we can do that. If you want them traceable, we can do that. If you want (and this is the clever one) them to be anonymous, yet irrefutable, we can even do that.
Book plug: Best text out there is Stefan Brands' Rethinking Public Key Infrastructures
Of course you're right - they're easily crackable. Some pointy-hair will misunderstand it, or an overworked contractor in a bank won't be allowed time to do it right, or (FFS!) someone will let M$oft implement any of it. It will be broken, and it will be some stupid trivial hole that does it.
Microsoft may not be very responsive to public opinion, but the British Government sure is.
HMG is pretty responsive to "public opinion", but it's the opinion that the editor of the Sun tells them to have.
Issues ? What are they ? William Hague's new hairdo - now that's an "issue" that's getting the haeadlines today.
Democratic governments have to provide equal access to government services
Why ? Joe Sixpack just doesn't care enough. Our election is going to be fought on the basis (as ever) of tax handouts, and if we're lucky, who's to blame for Foot and Mouth (a bunch of subsidy-scavenging sheep farmers in Cumbria, doing dodgy midnight deals with black sheep).
they risk having a large group of pissed off voters on election day.
How about pissed off IT geeks ? Three times in the last year this government has put me out of business. First IR35 made freelance IT contractors untenable (although it's still OK to work in just the same way if you're a lawyer or an accountant). Then RIP offered me a two year prison sentence for most interesting uses of one-shot crypto (I was just trying to sell music on-line, with a per-play charging model). Now they've lumped all IT "security" people in with nightclub bouncers and parking wheelclampers, so that I can't even login as root unless I've bought a licence and had a police background check. Again, lawyers and accountants serving the same industry have been exempted.
Blair wants 100% of all government services online by 2005.
Blair is still at the "Every school should have its own Information Superhighway" level of understanding. He's so clueless that yesterday he even let M$oft use him as a PR bunny for an Office XP launch photo-op.
An interplanetary spaceship called "ICANN" ?
Is this some sort of B Ark ?
Just as we start to build the first city-block trashing mechasaurs, you want to make them as smart as one of Cthulthu's grandchildren ? Isn't a big dumb dinosaur scarey enough for you ?
Never trust anything with tentacles and no backbone.