It looks like companies, specifically Adobe, are realizing that people want to switch from windows to linux
No. In Adobe's actual paying customers use Windows and Macs, in that order (unlike the/. horde who complain they would use Photoshop _if_ their favorite distro was supported, and less than 1% of them would pay for it even then) and are generally productive with their OS of choice because Windows 2000 and above, believe it or not, make for excellent desktop OSes.
Adobe's newfound love of Linux has nothing to do with what their customers want today -- they have realized that Microsoft is testing the waters of competing with Adobe on its own turf by offering design tools for Windows (sparkle, xaml and particularly xaml/anywhere, metro). Although the products are nowhere as polished as Adobe's and the audience is slightly different, Adobe has partnered with MS long enough to see the writing on the wall. It knows that it needs to hedge its bets by adopting and evangelizing alternate platforms (assuming the MS of the future is just as aggressive as today's about/not/ developing for Linux).
> Not because we're a basketcase, but because the west screwed us over.
The Colonial West screwed a lot of people back in the day, Japan, China, SE Asia and India included. Why are these countries not basket-cases as well? Maybe because they've learnt to get out of the victim mentality, get off aid and you know, actually develop their infrastructure.
Not having corrupt dictators/leaders (at least none as extravagantly corrupt as some African leaders *cough* Zimbabwe) probably helped too.
If you really want to be taken seriously, start by demanding an end to charity and start pushing public opinion on things that really matter, like farm subsidies.
Great point. The germs->scary link the media plays up is ridiculous. Hygiene is important but you can carry it too far, and when you've done that you've screwed up your immune system so bad that you need ever increasing levels of sanitization (and associated costs) *just to live*.
Makes you wonder how the millions living in unsanitary poverty stricken conditions around the world manage to out-do first-world growth rates.
> If it is programmed an equipped with appropriate sensors.. then yes.
Yes, sensors, especially basic ones, are easy. Now that you have sensors, can you imbue your toy with enough programming for it to have a survival instinct so that it can avoid certain things and seek out others, the goal being to build up enough of something that'd enable it to reproduce?
> I would say my fried she is not alive because she cannot have children
False Dichotomy. Your friend has all the apparatus to enable her to have children. She could have prepared herself socially for it too, by dating and/or marrying. That she cannot is because of unknown problems with her reproductive organs, not because she is not alive.
Pointing to vegetative, sterile or comatose humans and asking if they're alive is to create a false dilemma because the real question is not whether THEY are alive (that is a question for bioethicists -- sterile people display many other lifelike behaviors, so that's a no-brainer, many comatose people show signs of cognitive thought under EEGs, and then you have a stray case like non-cognitively alive Schiavo where the EEG and physical evidence shows her brain has decayed to the point that there's no measurable brain activity left but clearly her other organs are alive in the sense of response to stimuli -- ingesting food and air and producing waste CO2 etc) but rather whether the design on which they are based is capable of life. And we know by our very existence that the answer is yes.
And if you create a thousand of these 'toys' and even 80% don't manage to reproduce, that's okay... it'd show your basic idea was sound, you'd do better next time.
I think that it would soon be feasible to construct mechanical/electronical... device that could accomplish all things simple bacteria can.
The problem is that there is a huge gap between your thinking and actual practice. So far, the only place such toys have been built has been in computer simulations. If you indeed can build one of these with current technology, several people would be VERY interested. As it happens, our current level of expertise in nanoassembly places us years away from self-replicating nanoassemblies, which I believe are necessary for self-reproducing organisms.
Of course, as you say, response to stimuli + reproduction are thumb-rules that describe the most basic forms of life. However they are GOOD thumb rules because we know that more complex organisms can be derived from these simple ones. (In some ways I guess this is an unconscious hark to machine evolution, somewhat like biological evolution)
Nothing stops you from building something de novo that displays multiple high-order cognitive abilities (such as making a pretty watercolor and riding a bicycle) -- and if you do that I'm sure we'd revise our notions of life/non-life, but my intuition is that you'd find it very hard to accomplish.
High School Biology is simplified but that's because it's aimed at high-schoolers. But it's not as off as you imagine. response to stimuli AND reproduction are key traits: can your computer respond to a temperature or salinity change in a meaningful way? i.e., would it be able to sense danger to itself and move away? would it seek out nutrients, such as electricity? Further, can it reproduce sexually or asexually?
Programming in the response to stimuli is easy. Creating hardware which can do all of that AND reproduce sustainably isn't. Even if you make something limited (out of Lego bricks, say) that can do all of the above, you'll have created a very rudimentary form of life.
> defining life is not that easy [...] consciousness may be?
About consciousness-- I'll say one thing, I'm not a big fan of the term. I refuse to get into the a/theist wars but to venerate what you don't understand is a primitive and very human trait (early man worshipped fire and lightning, for example). Today you see that very same idea played out as Intelligent Design -- because we can't understand the origin of life, it must be the handiwork of God. On a less obvious level, people who believe in consciousness and 'soul' are similar: we currently do not understand how massively interconnected neural systems work AND we don't believe in God BUT we still need a pole on which to stick our uncertainties; hence, 'consciousness'/'soul'.
In fact from your statements about everything possessing a soul I speculate you believe in the classical Gaia hypothesis: that the Earth is 'alive' and a living organism. Again, I say: Gaia and Soul are every bit as bunk as ID is. Gaia's adherents do not understand how truly complex large-scale non-linear systems work AND they don't believe in God BUT need a pole to stick their uncertainties on, hence it helps them to think that the planet is a live organism in itself.
The scale has changed since floods, famine and disease was thought of as divine retribution, but like those even allegedly modern men think nothing of invoking mystical mumbo jumbo as a prop for their fears. Plus ça change.
> Who trades videos over P2P or buys video DVDs from Borders, Wal-Mart or Amazon?
dada21, you luddite, go check out torrentspy.com. The video trading scene mostly uses BitTorrent and eMule, and it is VERY vibrant. The only thing holding it back is that you need obscure codecs to play the files (not a problem on PCs but a real problem on devices). The situation with video today is EXACTLY like the situation after Napster went under.
As for who buys DVDs from Amazon etc -- lots of people do. And boxset sales have made sure that the portable DVD player has become a serious vacation accessory.
As for TV: 2006 is going to be the year of digital on-demand tv (streamed or streamed+cached). ADSL2 is getting more common, even in bandwidth backwaters like the UK (see previous BBC/digital TV story). Yahoo, Microsoft and Google are all getting into the game and more and more TV studios are seeing the writing on the wall -- their libraries of old programs can be endlessly monetized by streaming them over the internet. In a world where customers connect at 10Mbps+, it'd be stupid not to do it.
Video blogs and DIY video is probably the least important of the applications of a portable video player.
> Video iPod is trying to invent a market boom.
There was no 'boom' for MP3 players before the iPod either, unless you count geeks. iPod+iTMS made the MP3 player chic enough for Joe Public to want one. If Apple doesn't do the same for video, someone else will.
The only valid problem about portable video is that watching video takes up your full attention. But lots of people who play games on their cellphones on the subway -- I'm sure they wouldn't mind watching video.
Hey that's a great ide... Oh wait. If your laptop is lost or stolen or you drop it and break your hard drive, you're screwed.
Lots of folk (notably GMX) offer free, gigabyte sized SSLed WebDAV space. Laptop + WebDAV -> way to go. 1 Gig is a *lot* for standard office documents and code, and if it's not enough I'm sure you can set up a server with the space you need.
If I were running a company, I would fire that little ol' secretary or bean counter who couldn't be bothered to read the "don't click on attachments" policy.
Really? I bet they get loads of legitimate attachments every day they HAVE to open to do their job. Why don't you run server-side antivirus (and client-side antivirus too)? Or remove unsafe types (coupled with a mimetype sniff?) If I were a CIO I'd be thinking of firing YOU instead, your attitude of "it's all the user's fault" and refusal to think of alternatives is really quite unprofessional.
Here's a clue: most well-managed networks don't have virus problems anymore. The key word is 'well', which implies some expertise on the part of the IT person in charge, as opposed to a characteristic urge to vent on/.
No mercy, nothing. I'd fire them on the spot if they spread a worm throughout the company and shut down the mail servers by not following the policy.
If John Q Public didn't make computer mistakes, we could get by with far fewer IT people. Especially arrogant ones like you who somehow think they're running a lab in MIT.
Oh, and I like how firing people for making computer mistakes is modded Funny. Ha ha. Let's outsource these arrogant geeks to India. Ha ha. Equally funny?
Google's stock offering has been very carefully structured to give Larry and Sergei a lot more control[1] than the ordinary shareholding public (and besides, Google is still relatively closely held - the main shareholders of Google are its investors and they trust the founders+Eric Schmidt implicitly). In fact, the non-founding shareholders (mostly Kleiner Perkins and Sequoia) likely know that Google's 'goofiness' makes for good press and share price, and as long as the party continues they're unlikely to rock the boat.
Funny how everytime Microsoft comes out with a new version of Windows, the new version is "pretty stable", and the old one suddenly becomes the unstable mess
You mean like how every time Apple releases a product the older one becomes a PoS?:-p
Anyhow, if you think anyone's calling NT an 'unstable mess', you're smoking dope. I've been using NT since version 3.1 -- 3.1 was a tad slow (userland display drivers, mmm) but rock solid. NT4 was rock solid as well, except for one workstation with a bad display driver (S3 sent a patch for that one).
Windows 2000 was excellent as well but for device support (getting USB to work with NT4 was not a nice experience), ability to work with a wider hardware base, improved VXD model (minidrivers, I believe they're called) and a more modern shell (NT4 shipped with IE2 and the Win95-style explorer unless you applied the Shell Update with SP4 that upgraded you to IE4 and the new shell).
Similarly XP added even more hardware+media support, compatibility for a broader app base including games (important since the OS was going to be marketed to consumers as well).
And in all of this, stability on supported hardware has not suffered a bit. If anything, NT's ability to deal with poor hardware has improved over time, especially with XP.
Not true of NT 5.1 and 5.2 (XP, 2003). Most services run as 'Local Service' or 'Network Service' with differently grained privileges. System is still available for services that require it (including NT's crss and lsass processes).
Probably because Ubuntu encourages users to give away CDs (and the liveCD+installCD in one package thing is pure genius, it encourages people to give it a try). Get your own and spread them around.
> The average user of the future may not WANT to maintain his PC software environment in the face of constant security upgrades.
And of course, the user experience of 'constant security upgrades' will remain the same eternally, because Moses said so.
Jeez. Slashdot luddites. Start by checking out Firefox 1.5 and MSN Messenger 7.5, both of which use binary delta patches. Check out Vista, which can shut down subsystems to eliminate reboots in most cases (practically only _some_ device driver updates will cause reboots).
And honestly, for most users a 'Updates have been applied to your PC/Click here to restart' is not a big deal, certainly not a harrowing experience like some of the posters seem to suggest.
The fact that Joel Spolsky thinks that his niche of expertise qualifies them to pronounce judgement on everything in software is really, really amusing....
Actually, if you had reading comprehension skills, you would see that Spolsky was talking about shrinkwrap mass-market software; it was YOU (that is to say, the AC commenter) who was pronouncing judgements on 'everything'.
considering I've worked on solving OS level problems and I don't think working on Excel is quite the same. I suppose that argument is neither here nor there, though.
Considering that you are *repeatedly* bringing up your expertise but signing off as Anonymous Coward, why don't you paste a link to your CV so that we may know your credentials? Normally I wouldn't ask for this because I value online anonymity but if you're going to toot your own horn you'd better add _something_ to back yourself up.
> I've worked on solving OS level problems
By which should we understand you subscribe to LKML?
Having worked on a huge embedded software project (think switch) where we lifted the software up... in short, I don't think Joel knows what he's talking about.
This is almost as good as the AC who replied to Tom Christiansen asking "Do you even know anything about Perl?"
Comparing a embedded project (where newer revs of the software don't cause the capabilities of the hardware don't change) to an open-ended application (like a browser or a word-processor whose featureset is really limited by imagination) or a mass-market OS (which has to run on a wide range of hardware and take into account numerous idiosyncracies) is really comparing apples and oranges. The fact that Slashdotters think that their niche of expertise qualifies them to pronounce judgement on everything in software is really, really amusing.
And oh, about Spolsky, he is very clear about which niche his writings deal with.
And that, kids, is why you shouldn't take design advice from Slashdot. There's just too many armchair software purists around here whose experience solely consists of doing CS lab exercises or developing software solely for relatively smallscale in-house apps for their 'insights' to be broadly applicable to the real world.
As the Oddpost team blog has noted, they've been working on the new Yahoo mail interface ever since they were acquired. And yes, the new Yahoo Mail does work with Firefox.
I know a lot of people who would sigh with relief, happily accept a lightweight thin client and throw out that hideous, malware-ridden fat-client piece of junk in the corner that they never understood and rarely worked properly.
That's a lot of pejoratives in there, but you know, there are reasons we moved beyond mainframe-only.
a) the day they discover their DSL's down and they need to create a report/greeting card how hard do you think they're going to swear?
b) websites like Yahoo and MSN generate huge numbers of trouble tickets for relatively simple tasks like mail. Care to provision and run a support organization to take care of hundreds of thousands of consumers running your Office Suite?
c) People need mobile access to their data. Chips are getting faster (and now, cooler). Laptops, handhelds and phones are getting more powerful (and popular) every day. You mean people are going to buy those and then wait for ages as that 25MB RAW image of their cat or 800MB hi-def video of Junior at the ball game downloads s l o w l y from their online store?
Dream on, sucker.
And oh, about that 'rarely worked properly' thing: check out Microsoft OneCare. The fact that Microsoft is the vendor is irrelevant (Apple could easily offer something similar through mac.com, I believe they already offer some online antivirus), the point is that just because current OSes have crappy self-healing doesn't mean things will forever be like that -- the combination of a 'fat' client and a fat pipe can create some amazing stuff.
There was the NT 3.x TCP/IP stack, but that's less relevant because MS bought a 3rd party stack and bolted it to the OS (funny how/.-ers obsess about FTP.EXE when the whole darn stack was BSD-derived).
Then there was SFU, which actually shipped GNU tools, and MS even distributed source for the GNU tools they modified.
In my current job, we have dispensed with 'tools'. A datamodel is something you spec out into a requirements document, then just build straight into SQL tables. It might take a morning to build and test a table, with FK references, and initial cut at indexes.
Intensive batch jobs are ideal candidates for an in-DB solution like SQL or SProcs (or a thin layer of Pro*C). The problems come with complex systems.
Requirements change. Over time, new classes of users enter the system, with brand new demands. The same data store is used by hundreds of apps. Most of these apps aren't batch jobs. In this scenario if the apps are all talking to the database directly you have a recipe for disaster. In OOP terms, your DB is now public. The only way to bring sanity to development is to develop what is essentially an API for working with your data. That is what ORM can do, cleanly.
I feel your pain about putting queries into business logic, but to look at DLINQ as an ORM tool is to miss the point. In the real world lots of people embed SQL in code (because they don't have time/skills/resources to devote to ORM, or simply don't know any better, or whatever) and the code they churn out today is a mess, to put it charitably. Now these very people can use a far more robust system at very little extra cost.
>> Implementing DLINQ is really as simple as implmenting a pattern... and adding your own DB-specific code. > Adding the DB-specific code IS providing support.
I think you misunderstood me. When I said adding 'your' own DB-specific code, the generic 'you' was a DB-vendor. So someone from Oracle would have to implement the pattern and add Oracle-specific code, etc. The programmer who uses DLINQ to connect to an Oracle DB wouldn't have to do anything special other than add an a/r:OracleDLINQ.dll at compile time.
That said, C# 3.0 is in very early stages now, and of course I don't expect serious support from Oracle/IBM until it moves closer to beta.
Er, it might help if you actually read the spec. This isn't 'embedded SQL' in the sense of Pro*C -- the 'queries' are really a bit of (helpful) syntactic sugar over an object-oriented, typesafe set of expressions (you'll see the lambda expressions, new to C#, used heavily here):
from c in customers where c.City == "London" select c.Name
Of course, if you don't like it you can always pass strings of SQL text to the data layer, or do everything with StoredProcs -- after all, DLINQ helpfully shields you from ADO.NET but nothing stops you from using ADO.NET either directly or through alternate layers like NHibernate.
This should also answer your point about optimised SQL generation... the programmer does not type SQL into C#, SQL gen is still done in the background.
Also, I can't believe that MS C# is going to include support for MySQL, Postgresql etc, like Hibernate, NHibernate, JDO etc.
They don't have to. Implementing DLINQ is really as simple as implmenting a pattern (which Helsberg called the 'query expression pattern') and adding your own DB-specific code.
Currently Oracle and DB/2 ship libs for ADO.NET, you can be quite sure they'll ship libs for DLINQ. If the MySQL and Postgres communites want DLINQ support badly enough, I'm sure someone will write it.
Slashdot Mirror
It looks like companies, specifically Adobe, are realizing that people want to switch from windows to linux
/. horde who complain they would use Photoshop _if_ their favorite distro was supported, and less than 1% of them would pay for it even then) and are generally productive with their OS of choice because Windows 2000 and above, believe it or not, make for excellent desktop OSes.
/not/ developing for Linux).
No. In Adobe's actual paying customers use Windows and Macs, in that order (unlike the
Adobe's newfound love of Linux has nothing to do with what their customers want today -- they have realized that Microsoft is testing the waters of competing with Adobe on its own turf by offering design tools for Windows (sparkle, xaml and particularly xaml/anywhere, metro). Although the products are nowhere as polished as Adobe's and the audience is slightly different, Adobe has partnered with MS long enough to see the writing on the wall. It knows that it needs to hedge its bets by adopting and evangelizing alternate platforms (assuming the MS of the future is just as aggressive as today's about
> Not because we're a basketcase, but because the west screwed us over.
The Colonial West screwed a lot of people back in the day, Japan, China, SE Asia and India included. Why are these countries not basket-cases as well? Maybe because they've learnt to get out of the victim mentality, get off aid and you know, actually develop their infrastructure.
Not having corrupt dictators/leaders (at least none as extravagantly corrupt as some African leaders *cough* Zimbabwe) probably helped too.
If you really want to be taken seriously, start by demanding an end to charity and start pushing public opinion on things that really matter, like farm subsidies.
Great point. The germs->scary link the media plays up is ridiculous. Hygiene is important but you can carry it too far, and when you've done that you've screwed up your immune system so bad that you need ever increasing levels of sanitization (and associated costs) *just to live*.
Makes you wonder how the millions living in unsanitary poverty stricken conditions around the world manage to out-do first-world growth rates.
> If it is programmed an equipped with appropriate sensors.. then yes.
... it'd show your basic idea was sound, you'd do better next time.
Yes, sensors, especially basic ones, are easy. Now that you have sensors, can you imbue your toy with enough programming for it to have a survival instinct so that it can avoid certain things and seek out others, the goal being to build up enough of something that'd enable it to reproduce?
> I would say my fried she is not alive because she cannot have children
False Dichotomy. Your friend has all the apparatus to enable her to have children. She could have prepared herself socially for it too, by dating and/or marrying. That she cannot is because of unknown problems with her reproductive organs, not because she is not alive.
Pointing to vegetative, sterile or comatose humans and asking if they're alive is to create a false dilemma because the real question is not whether THEY are alive (that is a question for bioethicists -- sterile people display many other lifelike behaviors, so that's a no-brainer, many comatose people show signs of cognitive thought under EEGs, and then you have a stray case like non-cognitively alive Schiavo where the EEG and physical evidence shows her brain has decayed to the point that there's no measurable brain activity left but clearly her other organs are alive in the sense of response to stimuli -- ingesting food and air and producing waste CO2 etc) but rather whether the design on which they are based is capable of life. And we know by our very existence that the answer is yes.
And if you create a thousand of these 'toys' and even 80% don't manage to reproduce, that's okay
I think that it would soon be feasible to construct mechanical/electronical... device that could accomplish all things simple bacteria can.
The problem is that there is a huge gap between your thinking and actual practice. So far, the only place such toys have been built has been in computer simulations. If you indeed can build one of these with current technology, several people would be VERY interested. As it happens, our current level of expertise in nanoassembly places us years away from self-replicating nanoassemblies, which I believe are necessary for self-reproducing organisms.
Of course, as you say, response to stimuli + reproduction are thumb-rules that describe the most basic forms of life. However they are GOOD thumb rules because we know that more complex organisms can be derived from these simple ones. (In some ways I guess this is an unconscious hark to machine evolution, somewhat like biological evolution)
Nothing stops you from building something de novo that displays multiple high-order cognitive abilities (such as making a pretty watercolor and riding a bicycle) -- and if you do that I'm sure we'd revise our notions of life/non-life, but my intuition is that you'd find it very hard to accomplish.
High School Biology is simplified but that's because it's aimed at high-schoolers. But it's not as off as you imagine. response to stimuli AND reproduction are key traits: can your computer respond to a temperature or salinity change in a meaningful way? i.e., would it be able to sense danger to itself and move away? would it seek out nutrients, such as electricity? Further, can it reproduce sexually or asexually?
Programming in the response to stimuli is easy. Creating hardware which can do all of that AND reproduce sustainably isn't. Even if you make something limited (out of Lego bricks, say) that can do all of the above, you'll have created a very rudimentary form of life.
> defining life is not that easy [...] consciousness may be?
About consciousness-- I'll say one thing, I'm not a big fan of the term. I refuse to get into the a/theist wars but to venerate what you don't understand is a primitive and very human trait (early man worshipped fire and lightning, for example). Today you see that very same idea played out as Intelligent Design -- because we can't understand the origin of life, it must be the handiwork of God. On a less obvious level, people who believe in consciousness and 'soul' are similar: we currently do not understand how massively interconnected neural systems work AND we don't believe in God BUT we still need a pole on which to stick our uncertainties; hence, 'consciousness'/'soul'.
In fact from your statements about everything possessing a soul I speculate you believe in the classical Gaia hypothesis: that the Earth is 'alive' and a living organism. Again, I say: Gaia and Soul are every bit as bunk as ID is. Gaia's adherents do not understand how truly complex large-scale non-linear systems work AND they don't believe in God BUT need a pole to stick their uncertainties on, hence it helps them to think that the planet is a live organism in itself.
The scale has changed since floods, famine and disease was thought of as divine retribution, but like those even allegedly modern men think nothing of invoking mystical mumbo jumbo as a prop for their fears. Plus ça change.
> Who trades videos over P2P or buys video DVDs from Borders, Wal-Mart or Amazon?
dada21, you luddite, go check out torrentspy.com. The video trading scene mostly uses BitTorrent and eMule, and it is VERY vibrant. The only thing holding it back is that you need obscure codecs to play the files (not a problem on PCs but a real problem on devices). The situation with video today is EXACTLY like the situation after Napster went under.
As for who buys DVDs from Amazon etc -- lots of people do. And boxset sales have made sure that the portable DVD player has become a serious vacation accessory.
As for TV: 2006 is going to be the year of digital on-demand tv (streamed or streamed+cached). ADSL2 is getting more common, even in bandwidth backwaters like the UK (see previous BBC/digital TV story). Yahoo, Microsoft and Google are all getting into the game and more and more TV studios are seeing the writing on the wall -- their libraries of old programs can be endlessly monetized by streaming them over the internet. In a world where customers connect at 10Mbps+, it'd be stupid not to do it.
Video blogs and DIY video is probably the least important of the applications of a portable video player.
> Video iPod is trying to invent a market boom.
There was no 'boom' for MP3 players before the iPod either, unless you count geeks. iPod+iTMS made the MP3 player chic enough for Joe Public to want one. If Apple doesn't do the same for video, someone else will.
The only valid problem about portable video is that watching video takes up your full attention. But lots of people who play games on their cellphones on the subway -- I'm sure they wouldn't mind watching video.
> X-Windows
Um, you mean the X Window System. 5 digit UID-ers ought to set a good example, damnit.
Hey that's a great ide... Oh wait. If your laptop is lost or stolen or you drop it and break your hard drive, you're screwed.
Lots of folk (notably GMX) offer free, gigabyte sized SSLed WebDAV space. Laptop + WebDAV -> way to go. 1 Gig is a *lot* for standard office documents and code, and if it's not enough I'm sure you can set up a server with the space you need.
If I were running a company, I would fire that little ol' secretary or bean counter who couldn't be bothered to read the "don't click on attachments" policy.
/.
Really? I bet they get loads of legitimate attachments every day they HAVE to open to do their job. Why don't you run server-side antivirus (and client-side antivirus too)? Or remove unsafe types (coupled with a mimetype sniff?) If I were a CIO I'd be thinking of firing YOU instead, your attitude of "it's all the user's fault" and refusal to think of alternatives is really quite unprofessional.
Here's a clue: most well-managed networks don't have virus problems anymore. The key word is 'well', which implies some expertise on the part of the IT person in charge, as opposed to a characteristic urge to vent on
No mercy, nothing. I'd fire them on the spot if they spread a worm throughout the company and shut down the mail servers by not following the policy.
If John Q Public didn't make computer mistakes, we could get by with far fewer IT people. Especially arrogant ones like you who somehow think they're running a lab in MIT.
Oh, and I like how firing people for making computer mistakes is modded Funny. Ha ha. Let's outsource these arrogant geeks to India. Ha ha. Equally funny?
Google's stock offering has been very carefully structured to give Larry and Sergei a lot more control[1] than the ordinary shareholding public (and besides, Google is still relatively closely held - the main shareholders of Google are its investors and they trust the founders+Eric Schmidt implicitly). In fact, the non-founding shareholders (mostly Kleiner Perkins and Sequoia) likely know that Google's 'goofiness' makes for good press and share price, and as long as the party continues they're unlikely to rock the boat.
[1] WaPo: After IPO, Google Founders Plan to Remain in Control
Funny how everytime Microsoft comes out with a new version of Windows, the new version is "pretty stable", and the old one suddenly becomes the unstable mess
:-p
You mean like how every time Apple releases a product the older one becomes a PoS?
Anyhow, if you think anyone's calling NT an 'unstable mess', you're smoking dope.
I've been using NT since version 3.1 -- 3.1 was a tad slow (userland display drivers, mmm) but rock solid. NT4 was rock solid as well, except for one workstation with a bad display driver (S3 sent a patch for that one).
Windows 2000 was excellent as well but for device support (getting USB to work with NT4 was not a nice experience), ability to work with a wider hardware base, improved VXD model (minidrivers, I believe they're called) and a more modern shell (NT4 shipped with IE2 and the Win95-style explorer unless you applied the Shell Update with SP4 that upgraded you to IE4 and the new shell).
Similarly XP added even more hardware+media support, compatibility for a broader app base including games (important since the OS was going to be marketed to consumers as well).
And in all of this, stability on supported hardware has not suffered a bit. If anything, NT's ability to deal with poor hardware has improved over time, especially with XP.
> it's also the default user for Windows Services
Not true of NT 5.1 and 5.2 (XP, 2003). Most services run as 'Local Service' or 'Network Service' with differently grained privileges. System is still available for services that require it (including NT's crss and lsass processes).
Probably because Ubuntu encourages users to give away CDs (and the liveCD+installCD in one package thing is pure genius, it encourages people to give it a try). Get your own and spread them around.
> The average user of the future may not WANT to maintain his PC software environment in the face of constant security upgrades.
And of course, the user experience of 'constant security upgrades' will remain the same eternally, because Moses said so.
Jeez. Slashdot luddites. Start by checking out Firefox 1.5 and MSN Messenger 7.5, both of which use binary delta patches. Check out Vista, which can shut down subsystems to eliminate reboots in most cases (practically only _some_ device driver updates will cause reboots).
And honestly, for most users a 'Updates have been applied to your PC/Click here to restart' is not a big deal, certainly not a harrowing experience like some of the posters seem to suggest.
The fact that Joel Spolsky thinks that his niche of expertise qualifies them to pronounce judgement on everything in software is really, really amusing....
Actually, if you had reading comprehension skills, you would see that Spolsky was talking about shrinkwrap mass-market software; it was YOU (that is to say, the AC commenter) who was pronouncing judgements on 'everything'.
considering I've worked on solving OS level problems and I don't think working on Excel is quite the same. I suppose that argument is neither here nor there, though.
Considering that you are *repeatedly* bringing up your expertise but signing off as Anonymous Coward, why don't you paste a link to your CV so that we may know your credentials? Normally I wouldn't ask for this because I value online anonymity but if you're going to toot your own horn you'd better add _something_ to back yourself up.
> I've worked on solving OS level problems
By which should we understand you subscribe to LKML?
Having worked on a huge embedded software project (think switch) where we lifted the software up ... in short, I don't think Joel knows what he's talking about.
This is almost as good as the AC who replied to Tom Christiansen asking "Do you even know anything about Perl?"
Comparing a embedded project (where newer revs of the software don't cause the capabilities of the hardware don't change) to an open-ended application (like a browser or a word-processor whose featureset is really limited by imagination) or a mass-market OS (which has to run on a wide range of hardware and take into account numerous idiosyncracies) is really comparing apples and oranges. The fact that Slashdotters think that their niche of expertise qualifies them to pronounce judgement on everything in software is really, really amusing.
And oh, about Spolsky, he is very clear about which niche his writings deal with.
And that, kids, is why you shouldn't take design advice from Slashdot. There's just too many armchair software purists around here whose experience solely consists of doing CS lab exercises or developing software solely for relatively smallscale in-house apps for their 'insights' to be broadly applicable to the real world.
As the Oddpost team blog has noted, they've been working on the new Yahoo mail interface ever since they were acquired. And yes, the new Yahoo Mail does work with Firefox.
I know a lot of people who would sigh with relief, happily accept a lightweight thin client and throw out that hideous, malware-ridden fat-client piece of junk in the corner that they never understood and rarely worked properly.
That's a lot of pejoratives in there, but you know, there are reasons we moved beyond mainframe-only.
a) the day they discover their DSL's down and they need to create a report/greeting card how hard do you think they're going to swear?
b) websites like Yahoo and MSN generate huge numbers of trouble tickets for relatively simple tasks like mail. Care to provision and run a support organization to take care of hundreds of thousands of consumers running your Office Suite?
c) People need mobile access to their data. Chips are getting faster (and now, cooler). Laptops, handhelds and phones are getting more powerful (and popular) every day. You mean people are going to buy those and then wait for ages as that 25MB RAW image of their cat or 800MB hi-def video of Junior at the ball game downloads s l o w l y from their online store?
Dream on, sucker.
And oh, about that 'rarely worked properly' thing: check out Microsoft OneCare. The fact that Microsoft is the vendor is irrelevant (Apple could easily offer something similar through mac.com, I believe they already offer some online antivirus), the point is that just because current OSes have crappy self-healing doesn't mean things will forever be like that -- the combination of a 'fat' client and a fat pipe can create some amazing stuff.
> Is this an all-time low for a slashdot article?
In terms of general asininity I'd say that honor goes to this story.
Heh. It's actually the Compute Cluster Edition, but spelling and Slashdot go together like mercury and water...
There was the NT 3.x TCP/IP stack, but that's less relevant because MS bought a 3rd party stack and bolted it to the OS (funny how /.-ers obsess about FTP.EXE when the whole darn stack was BSD-derived).
Then there was SFU, which actually shipped GNU tools, and MS even distributed source for the GNU tools they modified.
In my current job, we have dispensed with 'tools'. A datamodel is something you spec out into a requirements document, then just build straight into SQL tables. It might take a morning to build and test a table, with FK references, and initial cut at indexes.
Intensive batch jobs are ideal candidates for an in-DB solution like SQL or SProcs (or a thin layer of Pro*C). The problems come with complex systems.
Requirements change. Over time, new classes of users enter the system, with brand new demands. The same data store is used by hundreds of apps. Most of these apps aren't batch jobs. In this scenario if the apps are all talking to the database directly you have a recipe for disaster. In OOP terms, your DB is now public. The only way to bring sanity to development is to develop what is essentially an API for working with your data. That is what ORM can do, cleanly.
I feel your pain about putting queries into business logic, but to look at DLINQ as an ORM tool is to miss the point. In the real world lots of people embed SQL in code (because they don't have time/skills/resources to devote to ORM, or simply don't know any better, or whatever) and the code they churn out today is a mess, to put it charitably. Now these very people can use a far more robust system at very little extra cost.
... and adding your own DB-specific code.
/r:OracleDLINQ.dll at compile time.
>> Implementing DLINQ is really as simple as implmenting a pattern
> Adding the DB-specific code IS providing support.
I think you misunderstood me. When I said adding 'your' own DB-specific code, the generic 'you' was a DB-vendor. So someone from Oracle would have to implement the pattern and add Oracle-specific code, etc. The programmer who uses DLINQ to connect to an Oracle DB wouldn't have to do anything special other than add an a
That said, C# 3.0 is in very early stages now, and of course I don't expect serious support from Oracle/IBM until it moves closer to beta.
turns into
Of course, if you don't like it you can always pass strings of SQL text to the data layer, or do everything with StoredProcs -- after all, DLINQ helpfully shields you from ADO.NET but nothing stops you from using ADO.NET either directly or through alternate layers like NHibernate.
This should also answer your point about optimised SQL generation
Also, I can't believe that MS C# is going to include support for MySQL, Postgresql etc, like Hibernate, NHibernate, JDO etc.
They don't have to. Implementing DLINQ is really as simple as implmenting a pattern (which Helsberg called the 'query expression pattern') and adding your own DB-specific code.
Currently Oracle and DB/2 ship libs for ADO.NET, you can be quite sure they'll ship libs for DLINQ. If the MySQL and Postgres communites want DLINQ support badly enough, I'm sure someone will write it.