> Sun botched the first applet plugin (which sucked). They rewrote it last year, which was recently released in an update.
Can you tell me where I get a Java plugin that doesn't suck? Because mine still does - it takes seconds to load, blocks the browser in the mean time, it always looks ugly (something wrong with the fonts?), and it often interferes with the web page. Plus the update mechanism is terrible - certainly if you have a normal user account for normal use.
Actually even the Flash plugin is a lot better, plus Flash graphics just look excellent.
> I'm sorry, but unless and until every browser has the "extensions" feature that FF has (Specifically including Adblock Plus and No Script) then NO browser will EVER be a true "Firefox Killer".
I agree that the extensions are great. I use ABP, NoScript and FlashBlock myself. However, I think it is worth noting that they do not always play along nicely, especially ABP and NoScript, and sometimes they also overlap functionality in the same area. That is of course the basic problem with extensions: you can never guarantee that all combinations will work. Sometimes fewer extensions and more built in features would be a benefit. NoScript and FlashBlock could very well be intergrated in Firefox, although ABP is probably too big and too changing.
And that only adds to the list of problems with Firefox: it is slow, memory hungry, single threaded, quite primitive out of the box, and not 100% standard compatible. Plugins are still a mess (although that may get better soon). Otherwise it is an excellent browser, but it is not flawless.
Chrome on the other hand is more stable, a lot faster, looks better, and brings more features out of the box. In fact it is pretty much perfect, if not for the lack of a few plugins: ABP, NoScript, and GreaseMonkey/Stylish.
> You don't run as administrator in Windows anymore, either.
And how many software packages still work then? Even Firefox had serious trouble with the update function under non-admin accounts until very recently.
> Security updates are likewise pushed in windows.
Pulled, to be precise. Via an Active-X plugin, yuck.
> Windows has an updating function.
No, it does not. The update "function" is a web site that sets of 3 security warnings in IE8.
> Your statements all show unfamiliarity with Windows.
> Java programs seem to launch pretty quickly on linux
I don't know - my browser always freezes for 3 to 4 seconds when a Java applet is launched - and MATLAB takes ages (15 seconds) to start, too, if you use the Java GUI. Given that good GUI feedback is instantaneous (up to the reaction time of 0.2 seconds), this means you can never write a good GUI in Java. Native apps take just a second or two - which is still a bit long, but ok.
Actually that is an important point. Most artificial light is so poor that it hinders whatever you try to do. It is well known how to produce better lighting, but it is just not done. When was the last time you experienced lighting that can adjust the color temperature, for example? That is quite an essential feature to keep your day rhythm working properly, and it has been shown to improve learning results significantly.
How are generics a good thing? They are a band aid for the fact that not everything in Java is an object. I would say that is the most serious defect of Java, apart from the lack of closures (yes, I know they are coming, but when?).
Plus any JVM I have seen is a piece of shit. Sorry, but if the official JVM takes several seconds just to start, that disqualifies it from a lot of perfectly good uses.
> Thanks. Due to Time Zones and all it nearly got me!
It's not just about time zones. Jokes have to be in the morning of the 1st of April, you have to make exactly 1, and they have to be funny. As a slashdot concession, you can just replace some value for funny, I guess, but the other rules stand!
Maybe so, but it seem to recall that it also included a fatal flaw for IPv6 - something along the line of not supporting DHCP for IPV6 or so. Can you imagine having to type in the local IPv6 address, the gateway and the DNS server? That would take a while!
> "Hmm... that's not such a big improvement. Maybe the best value is at 5!"
And that could well be correct - it is called the law of diminishing return. Doubling your fuel economy is great, doubling it again is more difficult, but less beneficial.
> Dozens of customers complaining daily and it took them three months to finally figure out "gee the whole block is down, let's go look at the router for this block."
That's what you get if you streamline your business by only hiring phone droids and cable monkeys. They are not paid to think, so they don't.
key-auth should be sufficient. The probability of guessing user name and the key is so absolutely remotely that 1 million login attempts do not get you any closer to it.
If the attempts are a performance problem (log - sync), the easiest option is to change the port, and the next is to restrict access with fail2ban or whatever flavor you fancy. Any obscure scheme will do, including "ping first with a payload of YEAH before ssh", just do not mistake obscurity for security.
> It seems everyone wants to be a "software engineer"
If only that were true. Engineering is a well established discipline focused on getting the job done properly. I don't think modern software can be explained in these terms at all.
The more appropriate term is "value engineering" (good enough is good enough), which is not an engineering term at all, but a business concept. The new goal is to whack something together as fast as possible that appears to kind of work on superficial inspection.
Of course that has nothing to with programming nor with engineering - Donald Knuth is certainly right about that.
> 20W 3dB 6 meters the power density is 0.0086 mW/cm2
I tend to agree - most likely there is very little radiation actually reaching the apartment. Just get a cheap RF meter and check how significant the background radiation is compared to a mobile phone. Chances are that it is negligible - and that would answer your question.
> Users want to keep doing whatever it is they are trying to do.
Exactly. And pop-up messages are annoying, pop-up errors doubly so. And on top of that, in my experience most error message are wrong, in that they may not give the actual reason for a failure, and they offer no alternative course of action.
"Transaction error: data could not be saved. Please try again later. [OK]" - Seems pretty good, hm? No, it is terrible, because it fails to answer the essential questions: where is the problem? is someone working on it? how long does it take? how should I continue?
Make your error messages helpful, and users will appreciate them. Make it say "1 entry stored locally until server becomes available, scheduled for 11:00am." on the top of the next page.
> Perhaps this means it'll get signal where I live?
Unfortunately now. All the energy going into your head is not going to reach the base station, so this is going to hurt your reception. It is just bad engineering, and they should really try to do better. This is not rocket science - the equations are very well understood and design tools are readily available.
You mean Uranium 235? There is still Uranium 238, which is very common. Then there Thorium, which is available just about everywhere. Chances are that we melt the earth before running out of Thorium, but if you still need more energy, there is always fusion.
The sources are not the problem, the sinks are. What do you do with radioactive waste? And how do you tell the public that everything is safe when the nuclear industry is lying like Dick Cheney? Those are the real questions.
Sorry, but the submitter got at wrong. A secret URL is essentially a password - so attempting lots of funny URLs can be like trying lots of ssh logins. The problem here is that it was a weak password, not that they used a secret URL.
Of course there are perfectly good reasons not to rely on secret URLs. Google is one, log files and browser caches are another. But that is a technical issue, and it is not actually relevant here.
> There's some fundamental flaw in the system if giving people free stuff is bad for them...
The flaw is that the free stuff will not flow forever. So it is important to maintain both commercially viable local systems, and a functioning local society. Both aspects have received way too little attention in Haiti, and that is the reason that most of the aid will fail in making a lasting impact.
> It's not a secret that lock-in was why IIS and IE were designed to complement each other.
Indeed. And the obvious solution is to use IE6 for legacy systems only, and Firefox/Chrome/Safari as the default browser for modern web sites. Problem solved. And when Windows 7 comes around, you can use the incredibly cumbersome XP compatibility machine.
> Current Flash-heavy sites do not work well on any other device either.
Is that a problem with Flash or with the site? Because I am sure that the Windows GUI, X11, and Javascript all have hover functionality. The key to use it only for hints, but not for essential functionality. This goes for any application in any language, if you want it to be useful on a touch screen device. Flash actually has very little to do with it.
> This is a opportunity that Microsoft clearly missed. Sure, they have Windows Update, which is not that bad and does the job for Windows itself, plus some drivers.
How do you conclude that it is "not that bad"? The idea is fine, but the implementation is one of the worst things Microsoft has delivered, and certainly the worst implementation that is widely used (due to a lack of alternatives).
> Very little software on Windows requires administrative privileges -- Vista forced those necessary fixes years ago.
If it did, it did not do it properly. Spotify, WoW, Firefox, Adobe Reader, Dreamweaver, Foxit... they all require administrative privileges to perform the (otherwise automatic) updates. Microsoft is doing a lot better: MSE works for regular users, as does the automatic update function.
> Oh wait, you wanted _Adobe_ to do something about it? I'm pretty sure they fired the only developer who understood their codebase years ago.
Expecting Adobe to write software is like expecting trees to migrate to a new habitat. It ain't going to happen. Adobe excels at design and marketing, and they want to keep it that way.
Slashdot Mirror
> Sun botched the first applet plugin (which sucked). They rewrote it last year, which was recently released in an update.
Can you tell me where I get a Java plugin that doesn't suck? Because mine still does - it takes seconds to load, blocks the browser in the mean time, it always looks ugly (something wrong with the fonts?), and it often interferes with the web page. Plus the update mechanism is terrible - certainly if you have a normal user account for normal use.
Actually even the Flash plugin is a lot better, plus Flash graphics just look excellent.
> I'm sorry, but unless and until every browser has the "extensions" feature that FF has (Specifically including Adblock Plus and No Script) then NO browser will EVER be a true "Firefox Killer".
I agree that the extensions are great. I use ABP, NoScript and FlashBlock myself. However, I think it is worth noting that they do not always play along nicely, especially ABP and NoScript, and sometimes they also overlap functionality in the same area. That is of course the basic problem with extensions: you can never guarantee that all combinations will work. Sometimes fewer extensions and more built in features would be a benefit. NoScript and FlashBlock could very well be intergrated in Firefox, although ABP is probably too big and too changing.
And that only adds to the list of problems with Firefox: it is slow, memory hungry, single threaded, quite primitive out of the box, and not 100% standard compatible. Plugins are still a mess (although that may get better soon). Otherwise it is an excellent browser, but it is not flawless.
Chrome on the other hand is more stable, a lot faster, looks better, and brings more features out of the box. In fact it is pretty much perfect, if not for the lack of a few plugins: ABP, NoScript, and GreaseMonkey/Stylish.
> You don't run as administrator in Windows anymore, either.
And how many software packages still work then? Even Firefox had serious trouble with the update function under non-admin accounts until very recently.
> Security updates are likewise pushed in windows.
Pulled, to be precise. Via an Active-X plugin, yuck.
> Windows has an updating function.
No, it does not. The update "function" is a web site that sets of 3 security warnings in IE8.
> Your statements all show unfamiliarity with Windows.
Ditto.
> Did Bing really fail?
I don't think so. It is not Google, but it is not bad either. So I think it is delivering pretty much as good as you could hope for.
> Remember when Microsoft introduced Internet Explorer, and put Netscape out of business?
Yes, and they nearly ruined the Internet? Even Microsoft noticed that this did not go all too well.
> Java programs seem to launch pretty quickly on linux
I don't know - my browser always freezes for 3 to 4 seconds when a Java applet is launched - and MATLAB takes ages (15 seconds) to start, too, if you use the Java GUI. Given that good GUI feedback is instantaneous (up to the reaction time of 0.2 seconds), this means you can never write a good GUI in Java. Native apps take just a second or two - which is still a bit long, but ok.
> Wow ... "increase the speed of learning?"
Actually that is an important point. Most artificial light is so poor that it hinders whatever you try to do. It is well known how to produce better lighting, but it is just not done. When was the last time you experienced lighting that can adjust the color temperature, for example? That is quite an essential feature to keep your day rhythm working properly, and it has been shown to improve learning results significantly.
> OTOH, Java has support for generics
How are generics a good thing? They are a band aid for the fact that not everything in Java is an object. I would say that is the most serious defect of Java, apart from the lack of closures (yes, I know they are coming, but when?).
Plus any JVM I have seen is a piece of shit. Sorry, but if the official JVM takes several seconds just to start, that disqualifies it from a lot of perfectly good uses.
> Thanks. Due to Time Zones and all it nearly got me!
It's not just about time zones. Jokes have to be in the morning of the 1st of April, you have to make exactly 1, and they have to be funny. As a slashdot concession, you can just replace some value for funny, I guess, but the other rules stand!
> apparently IPv6 will work on XP Pro SP1
Maybe so, but it seem to recall that it also included a fatal flaw for IPv6 - something along the line of not supporting DHCP for IPV6 or so. Can you imagine having to type in the local IPv6 address, the gateway and the DNS server? That would take a while!
> "Hmm... that's not such a big improvement. Maybe the best value is at 5!"
And that could well be correct - it is called the law of diminishing return. Doubling your fuel economy is great, doubling it again is more difficult, but less beneficial.
> Dozens of customers complaining daily and it took them three months to finally figure out "gee the whole block is down, let's go look at the router for this block."
That's what you get if you streamline your business by only hiring phone droids and cable monkeys. They are not paid to think, so they don't.
key-auth should be sufficient. The probability of guessing user name and the key is so absolutely remotely that 1 million login attempts do not get you any closer to it.
If the attempts are a performance problem (log - sync), the easiest option is to change the port, and the next is to restrict access with fail2ban or whatever flavor you fancy. Any obscure scheme will do, including "ping first with a payload of YEAH before ssh", just do not mistake obscurity for security.
> It seems everyone wants to be a "software engineer"
If only that were true. Engineering is a well established discipline focused on getting the job done properly. I don't think modern software can be explained in these terms at all.
The more appropriate term is "value engineering" (good enough is good enough), which is not an engineering term at all, but a business concept. The new goal is to whack something together as fast as possible that appears to kind of work on superficial inspection.
Of course that has nothing to with programming nor with engineering - Donald Knuth is certainly right about that.
> 20W 3dB 6 meters the power density is 0.0086 mW/cm2
I tend to agree - most likely there is very little radiation actually reaching the apartment. Just get a cheap RF meter and check how significant the background radiation is compared to a mobile phone. Chances are that it is negligible - and that would answer your question.
> Users want to keep doing whatever it is they are trying to do.
Exactly. And pop-up messages are annoying, pop-up errors doubly so. And on top of that, in my experience most error message are wrong, in that they may not give the actual reason for a failure, and they offer no alternative course of action.
"Transaction error: data could not be saved. Please try again later. [OK]" - Seems pretty good, hm? No, it is terrible, because it fails to answer the essential questions: where is the problem? is someone working on it? how long does it take? how should I continue?
Make your error messages helpful, and users will appreciate them. Make it say "1 entry stored locally until server becomes available, scheduled for 11:00am." on the top of the next page.
s/now/not/
> Perhaps this means it'll get signal where I live?
Unfortunately now. All the energy going into your head is not going to reach the base station, so this is going to hurt your reception. It is just bad engineering, and they should really try to do better. This is not rocket science - the equations are very well understood and design tools are readily available.
> What happens when the uranium runs out?
You mean Uranium 235? There is still Uranium 238, which is very common. Then there Thorium, which is available just about everywhere. Chances are that we melt the earth before running out of Thorium, but if you still need more energy, there is always fusion.
The sources are not the problem, the sinks are. What do you do with radioactive waste? And how do you tell the public that everything is safe when the nuclear industry is lying like Dick Cheney? Those are the real questions.
Sorry, but the submitter got at wrong. A secret URL is essentially a password - so attempting lots of funny URLs can be like trying lots of ssh logins. The problem here is that it was a weak password, not that they used a secret URL.
Of course there are perfectly good reasons not to rely on secret URLs. Google is one, log files and browser caches are another. But that is a technical issue, and it is not actually relevant here.
> There's some fundamental flaw in the system if giving people free stuff is bad for them...
The flaw is that the free stuff will not flow forever. So it is important to maintain both commercially viable local systems, and a functioning local society. Both aspects have received way too little attention in Haiti, and that is the reason that most of the aid will fail in making a lasting impact.
> It's not a secret that lock-in was why IIS and IE were designed to complement each other.
Indeed. And the obvious solution is to use IE6 for legacy systems only, and Firefox/Chrome/Safari as the default browser for modern web sites. Problem solved. And when Windows 7 comes around, you can use the incredibly cumbersome XP compatibility machine.
> Current Flash-heavy sites do not work well on any other device either.
Is that a problem with Flash or with the site? Because I am sure that the Windows GUI, X11, and Javascript all have hover functionality. The key to use it only for hints, but not for essential functionality. This goes for any application in any language, if you want it to be useful on a touch screen device. Flash actually has very little to do with it.
> This is a opportunity that Microsoft clearly missed. Sure, they have Windows Update, which is not that bad and does the job for Windows itself, plus some drivers.
How do you conclude that it is "not that bad"? The idea is fine, but the implementation is one of the worst things Microsoft has delivered, and certainly the worst implementation that is widely used (due to a lack of alternatives).
> Very little software on Windows requires administrative privileges -- Vista forced those necessary fixes years ago.
If it did, it did not do it properly. Spotify, WoW, Firefox, Adobe Reader, Dreamweaver, Foxit ... they all require administrative privileges to perform the (otherwise automatic) updates. Microsoft is doing a lot better: MSE works for regular users, as does the automatic update function.
> Oh wait, you wanted _Adobe_ to do something about it? I'm pretty sure they fired the only developer who understood their codebase years ago.
Expecting Adobe to write software is like expecting trees to migrate to a new habitat. It ain't going to happen. Adobe excels at design and marketing, and they want to keep it that way.