...if DVD's are killing the business, and driving you to financial ruin, then STOP SELLING THEM. Seems like he must be a pretty bad businessman to not understand that if something is hurting your bottom line, you should stop doing it. Maybe he came from the dot-com school of business; sell things at a loss but make it up in volume...
You can tweak some settings, but overall its like using the same film for everything.
This is not true at all. What makes various films different visually? A bunch of factors, some of the obvious ones are ability to hold detail in the shadows or highlights (dynamic range), contrast curves, sharpness (size and shape of grain), grain (size and shape of grain again), color balance, etc. All of these effects, except for grain are easily achievable with some post-processing. Want a Fuji NPH look? Desaturate a bit and compress the contrast. Want Velvia? Boost the saturation until grass looks neon and people look like turnips. It's all within reach, which is the flexibility of digital. Since your sensor is basically the same every time, you can even save the combination of adjustments and apply them on a repeated basis to entire series of shots at once.
Don't get me wrong, digital is expensive to do well at the moment. You're totally right about it costing $2200 or more to match the results of a $350 film camera (why I'm still shooting film primarily), but don't anyone make the mistake of thinking digital can't match film in flexibility. The digital darkroom is a different, but incredibly expressive environment, no matter how hard the purists argue that it's not artistic enough, because it uses bits and bytes instead of chemicals and emulsions.
The problem with using refreshes for missing or moved ocntent is precisely because it's transparent to the user. They aren't necessarily aware that the content is even moved, so the bookmark has no chance of getting changed, and you'll have to maintain those pages forever. Far better to 1) have a scheme to keep the URL's working (ideal), or 2) apoligize to the user for ruining their web experience and tell them were to find the new stuff, but don't redirect them automatically. That way they're aware that things have changed and have a (albiet slim) chance of updating their bookmark.
One other good use of refresh is when performing a long transaction to display a "please wait" type screen that updates the user as to progress.
It was tough to find the questions buried in that long commentary, but here are the answers:
1. Who pays for the factory retrofits?
It's not like we're talking about a completely new way of generating images that makes all old factories completely obsolete. Most of the parts stay the same, there are a few new ones, and some need to be built with better specs. The manufacturers are constantly upgrading their facilities, because it enables them to sell more TV's cheaper. So they'll pay, and if it does cost a lot, they'll pass some of it on to the consumer, just like they always do, but I'd be suprised if there's a huge jump in priace in 5 years. HDTV Tuners started out at $1000 and have dropped to $500 in about two years, and they're still not really being manufacturered in quantities yet. Plus, tuners to upgrade old sets don't need to be high-res, which makes them much, much less expensive.
2. Where do all the old TV sets go?
Where do they go now? Into landfills. That problem has got nothing to do with HDTV. There will be an easy, cheap, external box to buy that will make your TV continue working. Hollywood doesn't care about low-res, NTSC analog outputs, only the higher-resolution ones, so they don't need a "trusted chain".
There are things to be worried about for when it comes to freedom of digital media going forward, but this isn't one of them.
1. You can't get familiar with a PIX by using a free firewall, so it has some educational benefit (although if you "get" what firewalls do, the rest is mostly just syntax).
2. Stateful failover. I don't think any of the free options support this. With the PIX, you can plug two in via a serial cable in a master/slave configuration, and the master constantly sends it's state to the slave. If the master dies, the slave takes over and no TCP sessions are dropped. Only you can decide if this feature is important to you.
...of how businesses will misuse software to nullify it's potential producitivity benefit. The only situation I've ever seen Exchange used where it makes sense is in a small (
Beyond that, installations tend to be incredibly expensive, somewhat unreliable, and more importantly, usually pointless. It's not that Exchange necessarily is bad software, it's that it is nearly always horribly misued in large companies. Its primary purpose always breaks down to mail. The CEO types demand it because of "calendaring", "collaboration", and "advanced email features." They spend millions on hardware, software, and deployment/maintenance costs over a simpler email server to get those features...
Then they never use them. In big companies, the public folders are a wasteland of seldom used heirarchies that quickly become totally obsolete as the the organizational structure changes. Shared calendaring is a joke, that is barely more effecient than just e-mailing people with meeting notifications. Generally I have found that people sending out meeting notifications ignore the fact that their recipients have those areas blocked off on their calendar and send the meeting anyway, leaving the conflicts to be resolved via email anyway. Some companies get this right, but it could be done just as well with a non-integrated calendaring package (like Meeting Maker or something) that supports multiple platforms and is much, much less expensive. Finally advanced email features are almost never used. Encryption is spottily deployed if at all (and is almost inevitably flawed from a securitty standpoint). The address book is useful, but tends to end up as just yet another address book, leaving people multiple options to find things which is confusing (hopefully businesses will implement AD correctly and help alleviate this problem). Again, nothing a standalone LDAP server couldn't do better, faster, and cheaper.
Anyway, the long winded point is that like all enterprise software, everything except the very core functionality (email and sometimes shared calendaring) becomes corporate abaondware very quickly. Again, this is not necessarily Exchange's fault; it is the nature of the corporate beast. CEO's buy for features, then undertrained MSCE's roll out the software to thousands of people, occasionally doing lots of customization, occasionally with millions of dollars in consultant help. Of course the project has hundreds of departments to take input from, no real requirements, and a constantly changing business structure to map to. Tough to build something really useful with those contraints.
Yes, but they also said they would be blocking the RIAA's website for their customers, presumably so the RIAA can't exploit a browser bug to infect their systems.
Plus, it's not like they can't use one netblock to scan the P2P network, then a totally different one to try and take down the copyright infringing systems, totally avoiding the "honeynet."
While I admit, an action like this has an appeal in a very gut feeling kind of way, as a method of striking back against an unfair, almost unstoppable opponent, if we really sit back and think about it it's not a very workable solution. First off, it's censorship. We may not like the message, but ISP's blocking particular groups for their beliefs or acitivities is a pretty slippery slope. It's not so exciting if AOL decides to ban "hacker" groups from accessing AOL (or AOL customers from accessing them) because of the content (or activities) involved. Policy based filtering creates a fragmeted network, where company A's network can't reach customer B, etc.
Two, it's unworkable. Certainly whoever the MPAA/RIAA will employ to do their dirty work will be smart enough to come from a different location once they realize their IP's are blocked. This solution would be a little better if they had a system in place to block IP's that were actively attacking a P2P network, or a client, or a browser, etc. I see no problem with temporarily blocking small netblocks or IP's to stop an attack.
Remember; we'd cry bloody murder if the situation was reversed, and we'd be right to do so.
This is the most intelligent thing I've ever heard on slashdot before. If you don't understand this comment, read it again and again until you do.:-)
If you're a business, your money is far better spent improving the user experience rather than working on buying redundant-everything, building the support infrastructure, and incurring the extra overhead of the tedious and careful processes needed to obtain 5 nines (and 4, and even to a degeree 3 nines).
If your site sucks and no one visits, it doesn't really matter if it's down...work on building something reasonably reliable that is very compelling to your users; that's money much better spent...
I like this quote from Shutterfly about the issue, blaming browsers:
What we want to do is write once and have it work with everything," said Russ Sanon, senior manager for quality-assurance engineering at Shutterfly. "But it falls onto the lap of the individual browser manufacturer. There's nothing that we do that's proprietary. Everything that we write should work with W3C-complaint specs.
Funny. According to an html validator, I would put their site in the not
even close department. I wonder how the QA manager could claim his page is standards compliant, when the front page is so obviously not (although it would be a lot closer if it at least had the right DTD...). Could be because he doesn't understand what we are even talking about? Sigh.
Shareholders are just one form of money people. Google's accepted several million dollars (something like 50 I think; it's in the last article slashdot listed about Google's history) from venture capital firms; these are basically shareholders. Unless you're doing it yourself, with your own money and a small group of people you trust, there's always a trail of influence leading to the money.
How do you avoid business pressures to make short-sighted solutions, and consistently make good, common sense ideas work instead of adopting ones from marketing sources? Not only does Google have the best search engine technology, but you consistently do the "right" thing. Clean, quick homepage, text only well-identified ads, interesting research projects, etc...This is the way many search engines start, but they all went the way of the "dark" side instead of adopting the "right" solution. In my jobs, it's been very difficult to execute and justify good engineering (or just common sense) under pressure from the people who control the money. Any advice for driving through well-thought-out decisions instead of adopting the "management fad of the month"?
Bingo. I think (and hope personally) that this will drop us into a two format marketplace, just as it was before with Laserdiscs and Plain 'Ole VHS (POVHS?). And even more hopefully, it will light a fire under the DVD group to develop a real, honest to goodness HD-DVD format, instead of the MPEG-4, overly compressed stopgap that they are considering. D-Theater for the high-end and DVD for the low end, at least for a while; I can live with that. I won't be thrilled about the price of D-Theater (back to the LD price point), but at least my HDTV will be getting a workout. There's no reason these formats can't coexist, at least until an HD-DVD challenger arises.
Remember, thousands of people purchased LD players and discs, despite their higher cost, only marginally improved video (although much better audio), slow start, and the limited releases. With D-Theater offering such a quantum leap in quality, I can't really see how it would possibly go away entirely.
I realize this is carrying the discussion far beyond what is necessary, but I think you're missing the point a little.
Number one, I don't care that much, especially about my own player. My "broken" DVD player sits in my bedroom getting barely used if at all; if I cared I'd get it fixed. My primary DVD player plays all discs I've found so far.
2nd, it's not easy for the average consumer to find out if a player implements the spec. The real spec is only available if you license it as a manufacturer, which costs thousands of dollars; not really in most people's budget, or area of technical expertise. There's pretty much no way for the average consumer (or even a knowledgeable one really )to check for spec compliance. I suppose if you wanted to drop a few hundred bucks one one of the testing discs, you could go to a store and spend an hour with the player to see if it worked. Again, this is far beyond what most people are willing (or should have) to do. I think it's pretty arrogant on the part of designers to assume the public will be willing to do this for a $100-300 purchase. The manual and brochure for every player is going to say they support all DVD features, regardless of truth or quality.
Yes, it bugs me when people expect perfect, flawless performance from something like a DVD player when they buy a no-name one for $50 instead of purchasing a decent one; obviously you're going to lose some functionality. But the fact is that most people won't accept that; if it says 'DVD Player' on it, then they're going to expect it to play all DVD discs flawlessly. Look at the first low end "Progressive" DVD Player from Pioneer (DV-434 I believe...). The only point of progressive scan is to improve image quality, yet it was implemented so badly that image quality was noticably reduced. Some people were mad, most never noticed.
The DVD spec is large and complicated, and is not just a checkbox sign off list. DVD's and their players are basically software, which means when making a disc or a player, all you can do is test the crap out of them. The software engineering techniques used are no better than anywhere else, so there are going to be defects that slip into products, whether they be spec interpertation errors, deliberate corner cutting, or implementation errors. Period.
So when making a disc with a target audience of 35 million installed customers (or whatever number it gets to), it is typically in the manufacturer's best interest to stay within "safe" parts of the spec (just as just playing to movie and putting a few audio tracks on it) that pretty much all players have been extremely well tested under in all circumstances. That helps ensure less support calls and angry customers, which means more profit.
Branching is pretty rare, so you've got a large possibility of it not playing on a lot of players, which means more people are going to take it back and you're going to make less money. If it's a release such as Star Wars targetted at everyone, then it's very possible the media spin would make it look like your fault and you'd be stuck in a recall situation (or at least a bad PR spot). Something like the Terminator Ultimate edition is targeted at a much smaller audience (since there is already another (or is it two) release of the movie that the public has purchased), so it's much easier to justify branching.
Bottom line is, the manufacturers want to generally keep it simple, to ensure maximum hardware compatibility, which will help maximize their profits. Some manufacturers like the cutting edge though and will deal with the problems, typically those have a smaller potential audience.
So all I'm saying is that we probably won't have a branching version of Star Wars for a variety of reasons (content mastering difficulties really be the primary one). If we do, great, as long as it works on most players. I'd prefer (and spend the money for) a seperate disc version though, because it would leave more room for extras and be guaranteed to play on just about anything.
My DVD player works fine. Doesn't change the fact that there are thousands of playes out there that don't, because of a buggy or incomplete implementation. That means if a company releases a seamless branching disc, they need to be ready for a lot of returns and support calls from angry customers. It's definitely the DVD player's manufacturer's fault, but most consumers aren't going to realize that right off the bat.
How well does your DVD player handle progammed pan and scan (putting the widescreen movie on a disc, along with a control track that tells the player where to zoom and pan to create a reasonable pan and scan version on the fly)? That's part of the spec, but I can't think of a single disc that has used it yet, so the players, if they implement it at all, have at least relatively untested implementations, that are probably very buggy.
Also cheaper players often skimp on some of the peices of the spec. Layer switches are supposed to be _completely_ seamless, but how many players reall achieve that? How many players implement flag reading 3:2 pull down correctly for progressive output? Almost none. Hell, my old Toshiba 3108 2nd gen player from so long ago that I've moved to my bedroom doesn't even play DVD's with DVD-ROM content on them; it just locks up immediately requring a reboot (via unplugging) to start going again. Since it's out of warranty, Toshiba wants me to ship it to them and spend $50 to upgrade the firmware.
This is not an unusal example; the DVD spec is pretty large and complex, and the chances of getting a full blown, good performing implementation in a $150-100 player are generally pretty low. Certainly it's not one of the features that people base their decisions on.
So all I'm sayin' is, as a software manufacturer, it's an issue for them. Perhaps something like "Star Wars" would be enough of a killer app for people to get really pissed that their player didn't work with it, and force manufacturers into fixing their buggy implmentations. Who knows...
Somebody please mod the parent up (or better yet, update the story). This comment is dead on; this only represents a tiny change in Solaris licensing. Check out this Sun link for the Solaris 7 pricing scheme (Solaris 8 is mostly the same).
There are a few problems with that though...one is that some players (especially earlier ones) handle seamless branching very badly, doing everything from small skips on the branch to freezes and pauses to totally crashing the player. Hopefully as branching becomes more common those problems will (mostly) dissolve. The bigger problem is that with the special editions, it wasn't just added scenes that could be branched around, there were lots and lots of visual elements inserted into scenes, so you'd need two complete versions of a lot of scenes on the disc, one with and one without. It would take up a lot of disc space, and would tax branching to its limit (exposing more bugs in players and you reach out of the "comfortable" regions of the DVD spec, bringing us back to the first problem). It would be a tough balance to put both versions on a disc and not overcompress the video.
ESB and ROTJ would probably be better suited to this, since there were less scenes with added visual elements, mostly just visual cleanups (which can stay) and added scenes. A New Hope is the hard one to do. I'd rather they just make a finaly, definitive release of the original version, with just restored film and soundtrack, given as much bandwidth as possible on the disc. That's something I'd buy in a heartbeat. Better yet, put it on D-Theater, but that's probably just wishful thinking.
My bad, forgot all about laserdiscs. I think it has something to do with all that disc switching while watching the CAV version (only 30 min per side) of the whole trilogy back to back before Episode I came out.:-)
Personally, I don't care what he does to his movie. The sad part for me is that all the work done on restoring footage and reworking original soundtracks is all thrown into these modified special editions. The _original_ films now only exist as worn VHS tapes. The first special editions were different; not necessarily better, but different. Any further rounds of modification will be different still.
It's sad that the original films have become lost to consumers, and presumably will stay lost. How about if he releases the original, restored (but unaltered) movies on DVD first. Then he's free to add Jar Jar, Natalie Portman, Samuel Jackson, Matt LeBlanc or whoever the hell else he wants to Episodes IV-VI.
If it was in ACM communications, it should also be in their digital library on www.acm.com. Although it is only open for subscribers. Something I'd recommend though; there's hundreds of papers and articles there and you can get lost reading for a very long time.
In terms of the real world, certificates relate very little to the skill/knowledge of the person who has them. There are people who are excellent and what they do, and have vendor certifications. There are excellent people without vendor certs. There are frighteningly incompetent people with certs, and there are frighteningly incompetent people without certs. There are also all ranges in between. To boot, in my experience, there really isn't even a useful average quality behind certifications, so selecting a "certified" employee doesn't even increase your chances of getting someone with skill.
In short, there is no correlation that I've found between certs and quality of employees. Sadly, out in the world, some places take them very seriously. Some places they will just be your foot in the door, somce places they will get you the job, and some places won't even look at them; it really just depends. In a perfect world, I wouldn't even work for a place that required them or even took them into account since they are so suspect as a yardstick. Of course, it's the real world, and you have to work to eat.
In the long-winded end, certs typically can't hurt you, but from what I've seen, I wouldn't waste a lot of money or time trying to get them. These days the market is tight, so recruiters tend to not even talk to people who don't match every single requirement for a position, and those often include certs. If you're working with the end employer, you can often talk to the technical person and help him/her see past not having a cert. With an HR person or recruiter who isn't familiar with the subject matter, you'll probably need that cert. YMMV, but if I were doing the hiring, I'd never even mention/look at certifications.
Yeah. If we had something like that, then we would cut down on the number of unskilled programmers out there. Then we would have less security problems and less software failures. That would be a disaster.
Just like not everyone is qualified to design a bridge, and therefore not allowed to, not everyone is qualified to be a programmer, no matter how hard they try or no matter how many "Learn C in 60 days" books they read. If you're programming something like Word, then it probably doesn't matter that much. But operating systems (which have a habit of finding their way into mission critical systems) and custom large system programming (such as shuttle software, medical software, banking software, etc) should not be coded by fresh-out-of-school kids with no experience, or retrained truck drivers with 6 months of book reading behind them (no offense to truck drivers).
Truthfully, I don't know that certification or "oaths" are necesssarily the answer, but the market is apparently not doing it's job here, because it's extremely profitable to build shoddy software.
And perhaps not even a pure misapplication. Since there wasn't really a system designed and deployed yet to shoot down missiles, and you've got missles flying at your troops and friendly cities, but you have a system that's very close to the mission you need...
Why not give it a shot. So it didn't lend itself well to the purpose, and it was way out of the original design spec which caused major problems with it's effectiveness. However, if I'm sitting around in the bunker, I'd rather have someone trying to shoot down an incoming missile then sitting on their ass thinking "Damn, too bad this missile system was designed to shoot down airplanes and not missiles."
Of course, battlefields are notoriously improvisational environments, so it pays to add as much flexibility and robustness to your designs as possible...
Slashdot Mirror
...if DVD's are killing the business, and driving you to financial ruin, then STOP SELLING THEM. Seems like he must be a pretty bad businessman to not understand that if something is hurting your bottom line, you should stop doing it. Maybe he came from the dot-com school of business; sell things at a loss but make it up in volume...
...which makes it very difficult to crack into and steal its information...
You can tweak some settings, but overall its like using the same film for everything.
This is not true at all. What makes various films different visually? A bunch of factors, some of the obvious ones are ability to hold detail in the shadows or highlights (dynamic range), contrast curves, sharpness (size and shape of grain), grain (size and shape of grain again), color balance, etc. All of these effects, except for grain are easily achievable with some post-processing. Want a Fuji NPH look? Desaturate a bit and compress the contrast. Want Velvia? Boost the saturation until grass looks neon and people look like turnips. It's all within reach, which is the flexibility of digital. Since your sensor is basically the same every time, you can even save the combination of adjustments and apply them on a repeated basis to entire series of shots at once.
Don't get me wrong, digital is expensive to do well at the moment. You're totally right about it costing $2200 or more to match the results of a $350 film camera (why I'm still shooting film primarily), but don't anyone make the mistake of thinking digital can't match film in flexibility. The digital darkroom is a different, but incredibly expressive environment, no matter how hard the purists argue that it's not artistic enough, because it uses bits and bytes instead of chemicals and emulsions.
The problem with using refreshes for missing or moved ocntent is precisely because it's transparent to the user. They aren't necessarily aware that the content is even moved, so the bookmark has no chance of getting changed, and you'll have to maintain those pages forever. Far better to 1) have a scheme to keep the URL's working (ideal), or 2) apoligize to the user for ruining their web experience and tell them were to find the new stuff, but don't redirect them automatically. That way they're aware that things have changed and have a (albiet slim) chance of updating their bookmark.
One other good use of refresh is when performing a long transaction to display a "please wait" type screen that updates the user as to progress.
It was tough to find the questions buried in that long commentary, but here are the answers:
1. Who pays for the factory retrofits?
It's not like we're talking about a completely new way of generating images that makes all old factories completely obsolete. Most of the parts stay the same, there are a few new ones, and some need to be built with better specs. The manufacturers are constantly upgrading their facilities, because it enables them to sell more TV's cheaper. So they'll pay, and if it does cost a lot, they'll pass some of it on to the consumer, just like they always do, but I'd be suprised if there's a huge jump in priace in 5 years. HDTV Tuners started out at $1000 and have dropped to $500 in about two years, and they're still not really being manufacturered in quantities yet. Plus, tuners to upgrade old sets don't need to be high-res, which makes them much, much less expensive.
2. Where do all the old TV sets go?
Where do they go now? Into landfills. That problem has got nothing to do with HDTV. There will be an easy, cheap, external box to buy that will make your TV continue working. Hollywood doesn't care about low-res, NTSC analog outputs, only the higher-resolution ones, so they don't need a "trusted chain".
There are things to be worried about for when it comes to freedom of digital media going forward, but this isn't one of them.
1. You can't get familiar with a PIX by using a free firewall, so it has some educational benefit (although if you "get" what firewalls do, the rest is mostly just syntax).
2. Stateful failover. I don't think any of the free options support this. With the PIX, you can plug two in via a serial cable in a master/slave configuration, and the master constantly sends it's state to the slave. If the master dies, the slave takes over and no TCP sessions are dropped. Only you can decide if this feature is important to you.
...of how businesses will misuse software to nullify it's potential producitivity benefit. The only situation I've ever seen Exchange used where it makes sense is in a small (
Beyond that, installations tend to be incredibly expensive, somewhat unreliable, and more importantly, usually pointless. It's not that Exchange necessarily is bad software, it's that it is nearly always horribly misued in large companies. Its primary purpose always breaks down to mail. The CEO types demand it because of "calendaring", "collaboration", and "advanced email features." They spend millions on hardware, software, and deployment/maintenance costs over a simpler email server to get those features...
Then they never use them. In big companies, the public folders are a wasteland of seldom used heirarchies that quickly become totally obsolete as the the organizational structure changes. Shared calendaring is a joke, that is barely more effecient than just e-mailing people with meeting notifications. Generally I have found that people sending out meeting notifications ignore the fact that their recipients have those areas blocked off on their calendar and send the meeting anyway, leaving the conflicts to be resolved via email anyway. Some companies get this right, but it could be done just as well with a non-integrated calendaring package (like Meeting Maker or something) that supports multiple platforms and is much, much less expensive. Finally advanced email features are almost never used. Encryption is spottily deployed if at all (and is almost inevitably flawed from a securitty standpoint). The address book is useful, but tends to end up as just yet another address book, leaving people multiple options to find things which is confusing (hopefully businesses will implement AD correctly and help alleviate this problem). Again, nothing a standalone LDAP server couldn't do better, faster, and cheaper.
Anyway, the long winded point is that like all enterprise software, everything except the very core functionality (email and sometimes shared calendaring) becomes corporate abaondware very quickly. Again, this is not necessarily Exchange's fault; it is the nature of the corporate beast. CEO's buy for features, then undertrained MSCE's roll out the software to thousands of people, occasionally doing lots of customization, occasionally with millions of dollars in consultant help. Of course the project has hundreds of departments to take input from, no real requirements, and a constantly changing business structure to map to. Tough to build something really useful with those contraints.
Yes, but they also said they would be blocking the RIAA's website for their customers, presumably so the RIAA can't exploit a browser bug to infect their systems.
Plus, it's not like they can't use one netblock to scan the P2P network, then a totally different one to try and take down the copyright infringing systems, totally avoiding the "honeynet."
It's really just not a very useful idea.
While I admit, an action like this has an appeal in a very gut feeling kind of way, as a method of striking back against an unfair, almost unstoppable opponent, if we really sit back and think about it it's not a very workable solution. First off, it's censorship. We may not like the message, but ISP's blocking particular groups for their beliefs or acitivities is a pretty slippery slope. It's not so exciting if AOL decides to ban "hacker" groups from accessing AOL (or AOL customers from accessing them) because of the content (or activities) involved. Policy based filtering creates a fragmeted network, where company A's network can't reach customer B, etc.
Two, it's unworkable. Certainly whoever the MPAA/RIAA will employ to do their dirty work will be smart enough to come from a different location once they realize their IP's are blocked. This solution would be a little better if they had a system in place to block IP's that were actively attacking a P2P network, or a client, or a browser, etc. I see no problem with temporarily blocking small netblocks or IP's to stop an attack.
Remember; we'd cry bloody murder if the situation was reversed, and we'd be right to do so.
Until it's at zero, it can always go lower...:-)
This is the most intelligent thing I've ever heard on slashdot before. If you don't understand this comment, read it again and again until you do. :-)
If you're a business, your money is far better spent improving the user experience rather than working on buying redundant-everything, building the support infrastructure, and incurring the extra overhead of the tedious and careful processes needed to obtain 5 nines (and 4, and even to a degeree 3 nines).
If your site sucks and no one visits, it doesn't really matter if it's down...work on building something reasonably reliable that is very compelling to your users; that's money much better spent...
I like this quote from Shutterfly about the issue, blaming browsers:
What we want to do is write once and have it work with everything," said Russ Sanon, senior manager for quality-assurance engineering at Shutterfly. "But it falls onto the lap of the individual browser manufacturer. There's nothing that we do that's proprietary. Everything that we write should work with W3C-complaint specs.
Funny. According to an html validator, I would put their site in the not even close department. I wonder how the QA manager could claim his page is standards compliant, when the front page is so obviously not (although it would be a lot closer if it at least had the right DTD...). Could be because he doesn't understand what we are even talking about? Sigh.
Shareholders are just one form of money people. Google's accepted several million dollars (something like 50 I think; it's in the last article slashdot listed about Google's history) from venture capital firms; these are basically shareholders. Unless you're doing it yourself, with your own money and a small group of people you trust, there's always a trail of influence leading to the money.
How do you avoid business pressures to make short-sighted solutions, and consistently make good, common sense ideas work instead of adopting ones from marketing sources? Not only does Google have the best search engine technology, but you consistently do the "right" thing. Clean, quick homepage, text only well-identified ads, interesting research projects, etc...This is the way many search engines start, but they all went the way of the "dark" side instead of adopting the "right" solution. In my jobs, it's been very difficult to execute and justify good engineering (or just common sense) under pressure from the people who control the money. Any advice for driving through well-thought-out decisions instead of adopting the "management fad of the month"?
Bingo. I think (and hope personally) that this will drop us into a two format marketplace, just as it was before with Laserdiscs and Plain 'Ole VHS (POVHS?). And even more hopefully, it will light a fire under the DVD group to develop a real, honest to goodness HD-DVD format, instead of the MPEG-4, overly compressed stopgap that they are considering. D-Theater for the high-end and DVD for the low end, at least for a while; I can live with that. I won't be thrilled about the price of D-Theater (back to the LD price point), but at least my HDTV will be getting a workout. There's no reason these formats can't coexist, at least until an HD-DVD challenger arises.
Remember, thousands of people purchased LD players and discs, despite their higher cost, only marginally improved video (although much better audio), slow start, and the limited releases. With D-Theater offering such a quantum leap in quality, I can't really see how it would possibly go away entirely.
I realize this is carrying the discussion far beyond what is necessary, but I think you're missing the point a little.
:-)
Number one, I don't care that much, especially about my own player. My "broken" DVD player sits in my bedroom getting barely used if at all; if I cared I'd get it fixed. My primary DVD player plays all discs I've found so far.
2nd, it's not easy for the average consumer to find out if a player implements the spec. The real spec is only available if you license it as a manufacturer, which costs thousands of dollars; not really in most people's budget, or area of technical expertise. There's pretty much no way for the average consumer (or even a knowledgeable one really )to check for spec compliance. I suppose if you wanted to drop a few hundred bucks one one of the testing discs, you could go to a store and spend an hour with the player to see if it worked. Again, this is far beyond what most people are willing (or should have) to do. I think it's pretty arrogant on the part of designers to assume the public will be willing to do this for a $100-300 purchase. The manual and brochure for every player is going to say they support all DVD features, regardless of truth or quality.
Yes, it bugs me when people expect perfect, flawless performance from something like a DVD player when they buy a no-name one for $50 instead of purchasing a decent one; obviously you're going to lose some functionality. But the fact is that most people won't accept that; if it says 'DVD Player' on it, then they're going to expect it to play all DVD discs flawlessly. Look at the first low end "Progressive" DVD Player from Pioneer (DV-434 I believe...). The only point of progressive scan is to improve image quality, yet it was implemented so badly that image quality was noticably reduced. Some people were mad, most never noticed.
The DVD spec is large and complicated, and is not just a checkbox sign off list. DVD's and their players are basically software, which means when making a disc or a player, all you can do is test the crap out of them. The software engineering techniques used are no better than anywhere else, so there are going to be defects that slip into products, whether they be spec interpertation errors, deliberate corner cutting, or implementation errors. Period.
So when making a disc with a target audience of 35 million installed customers (or whatever number it gets to), it is typically in the manufacturer's best interest to stay within "safe" parts of the spec (just as just playing to movie and putting a few audio tracks on it) that pretty much all players have been extremely well tested under in all circumstances. That helps ensure less support calls and angry customers, which means more profit.
Branching is pretty rare, so you've got a large possibility of it not playing on a lot of players, which means more people are going to take it back and you're going to make less money. If it's a release such as Star Wars targetted at everyone, then it's very possible the media spin would make it look like your fault and you'd be stuck in a recall situation (or at least a bad PR spot). Something like the Terminator Ultimate edition is targeted at a much smaller audience (since there is already another (or is it two) release of the movie that the public has purchased), so it's much easier to justify branching.
Bottom line is, the manufacturers want to generally keep it simple, to ensure maximum hardware compatibility, which will help maximize their profits. Some manufacturers like the cutting edge though and will deal with the problems, typically those have a smaller potential audience.
So all I'm saying is that we probably won't have a branching version of Star Wars for a variety of reasons (content mastering difficulties really be the primary one). If we do, great, as long as it works on most players. I'd prefer (and spend the money for) a seperate disc version though, because it would leave more room for extras and be guaranteed to play on just about anything.
Okay. I'm done now.
My DVD player works fine. Doesn't change the fact that there are thousands of playes out there that don't, because of a buggy or incomplete implementation. That means if a company releases a seamless branching disc, they need to be ready for a lot of returns and support calls from angry customers. It's definitely the DVD player's manufacturer's fault, but most consumers aren't going to realize that right off the bat.
How well does your DVD player handle progammed pan and scan (putting the widescreen movie on a disc, along with a control track that tells the player where to zoom and pan to create a reasonable pan and scan version on the fly)? That's part of the spec, but I can't think of a single disc that has used it yet, so the players, if they implement it at all, have at least relatively untested implementations, that are probably very buggy.
Also cheaper players often skimp on some of the peices of the spec. Layer switches are supposed to be _completely_ seamless, but how many players reall achieve that? How many players implement flag reading 3:2 pull down correctly for progressive output? Almost none. Hell, my old Toshiba 3108 2nd gen player from so long ago that I've moved to my bedroom doesn't even play DVD's with DVD-ROM content on them; it just locks up immediately requring a reboot (via unplugging) to start going again. Since it's out of warranty, Toshiba wants me to ship it to them and spend $50 to upgrade the firmware.
This is not an unusal example; the DVD spec is pretty large and complex, and the chances of getting a full blown, good performing implementation in a $150-100 player are generally pretty low. Certainly it's not one of the features that people base their decisions on.
So all I'm sayin' is, as a software manufacturer, it's an issue for them. Perhaps something like "Star Wars" would be enough of a killer app for people to get really pissed that their player didn't work with it, and force manufacturers into fixing their buggy implmentations. Who knows...
Somebody please mod the parent up (or better yet, update the story). This comment is dead on; this only represents a tiny change in Solaris licensing. Check out this Sun link for the Solaris 7 pricing scheme (Solaris 8 is mostly the same).
There are a few problems with that though...one is that some players (especially earlier ones) handle seamless branching very badly, doing everything from small skips on the branch to freezes and pauses to totally crashing the player. Hopefully as branching becomes more common those problems will (mostly) dissolve. The bigger problem is that with the special editions, it wasn't just added scenes that could be branched around, there were lots and lots of visual elements inserted into scenes, so you'd need two complete versions of a lot of scenes on the disc, one with and one without. It would take up a lot of disc space, and would tax branching to its limit (exposing more bugs in players and you reach out of the "comfortable" regions of the DVD spec, bringing us back to the first problem). It would be a tough balance to put both versions on a disc and not overcompress the video.
ESB and ROTJ would probably be better suited to this, since there were less scenes with added visual elements, mostly just visual cleanups (which can stay) and added scenes. A New Hope is the hard one to do. I'd rather they just make a finaly, definitive release of the original version, with just restored film and soundtrack, given as much bandwidth as possible on the disc. That's something I'd buy in a heartbeat. Better yet, put it on D-Theater, but that's probably just wishful thinking.
My bad, forgot all about laserdiscs. I think it has something to do with all that disc switching while watching the CAV version (only 30 min per side) of the whole trilogy back to back before Episode I came out. :-)
Personally, I don't care what he does to his movie. The sad part for me is that all the work done on restoring footage and reworking original soundtracks is all thrown into these modified special editions. The _original_ films now only exist as worn VHS tapes. The first special editions were different; not necessarily better, but different. Any further rounds of modification will be different still.
It's sad that the original films have become lost to consumers, and presumably will stay lost. How about if he releases the original, restored (but unaltered) movies on DVD first. Then he's free to add Jar Jar, Natalie Portman, Samuel Jackson, Matt LeBlanc or whoever the hell else he wants to Episodes IV-VI.
If it was in ACM communications, it should also be in their digital library on www.acm.com. Although it is only open for subscribers. Something I'd recommend though; there's hundreds of papers and articles there and you can get lost reading for a very long time.
In terms of the real world, certificates relate very little to the skill/knowledge of the person who has them. There are people who are excellent and what they do, and have vendor certifications. There are excellent people without vendor certs. There are frighteningly incompetent people with certs, and there are frighteningly incompetent people without certs. There are also all ranges in between. To boot, in my experience, there really isn't even a useful average quality behind certifications, so selecting a "certified" employee doesn't even increase your chances of getting someone with skill.
In short, there is no correlation that I've found between certs and quality of employees. Sadly, out in the world, some places take them very seriously. Some places they will just be your foot in the door, somce places they will get you the job, and some places won't even look at them; it really just depends. In a perfect world, I wouldn't even work for a place that required them or even took them into account since they are so suspect as a yardstick. Of course, it's the real world, and you have to work to eat.
In the long-winded end, certs typically can't hurt you, but from what I've seen, I wouldn't waste a lot of money or time trying to get them. These days the market is tight, so recruiters tend to not even talk to people who don't match every single requirement for a position, and those often include certs. If you're working with the end employer, you can often talk to the technical person and help him/her see past not having a cert. With an HR person or recruiter who isn't familiar with the subject matter, you'll probably need that cert. YMMV, but if I were doing the hiring, I'd never even mention/look at certifications.
Yeah. If we had something like that, then we would cut down on the number of unskilled programmers out there. Then we would have less security problems and less software failures. That would be a disaster.
Just like not everyone is qualified to design a bridge, and therefore not allowed to, not everyone is qualified to be a programmer, no matter how hard they try or no matter how many "Learn C in 60 days" books they read. If you're programming something like Word, then it probably doesn't matter that much. But operating systems (which have a habit of finding their way into mission critical systems) and custom large system programming (such as shuttle software, medical software, banking software, etc) should not be coded by fresh-out-of-school kids with no experience, or retrained truck drivers with 6 months of book reading behind them (no offense to truck drivers).
Truthfully, I don't know that certification or "oaths" are necesssarily the answer, but the market is apparently not doing it's job here, because it's extremely profitable to build shoddy software.
And perhaps not even a pure misapplication. Since there wasn't really a system designed and deployed yet to shoot down missiles, and you've got missles flying at your troops and friendly cities, but you have a system that's very close to the mission you need...
Why not give it a shot. So it didn't lend itself well to the purpose, and it was way out of the original design spec which caused major problems with it's effectiveness. However, if I'm sitting around in the bunker, I'd rather have someone trying to shoot down an incoming missile then sitting on their ass thinking "Damn, too bad this missile system was designed to shoot down airplanes and not missiles."
Of course, battlefields are notoriously improvisational environments, so it pays to add as much flexibility and robustness to your designs as possible...