Likely, your neighbors won't be thinking twice about security INSIDE the network, since they "know" they are secure now, making it easier to browse their pr0n collection.
It's not exactly uncommon for even moderately security-aware people to have all kinds of open doors locally behind their lovely little NAT-box, making it very relatively easy for a malicious neighbour to cause all kinds of damage.
What are the chances you pissed some neighbour off in the past month?
Q: How are you going to steal my bytes when I don't pub my SSID? A: iwlist still shows it.
Uhm, is it possible you haven't realized by now, that "it" in the original answer does not refer to "my SSID", but to the network, as "my bytes"? I had no problems getting the point the first time, and I've seen no backpedaling.
Lovely "discussion", both of you managing to convey nothing in so many words. That's what I love about slashdot, keep the good work up!
That may take a while, at least for reaction-heavy games.
Ignoring latencies in active repeaters (routers, switches...) just the speed of light in fiber has a measurable roundtrip (about 12ms from Stockholm to Paris). Then add audio-video-encoding, which realistically needs to buffer at least a few frames to get decent compression (x264 for live-broadcast is usually varying from 500-1000ms), so that adds another 20-30ms. Then you'd either need strong FEC, or very fast re-transmission to not occasionally suffer heavy distortions in the picture (oops, there was a grenade), which either adds extra bandwidth and a little extra delay, or more delay with less bandwidth (could possibly be migigated with a failure-resilient wavelet codec). In the end, somewhere around 50ms latency is probably unavoidable, at which point all reaction-heavy games is a nono for many gamers. (Equal to ~20FPS)
To compete with graphical quality of a local game, you'd either need closer 100mbit/user to get even decent reaction-times using current and coming codecs, or a lot more for great reaction-times. That would of course be feasible for smaller edge-segments of the network, such as perhaps a gaming-server in the basement of every apartment-building, but that quickly becomes a logistics and service-nightmare.
Other than that, reaction-sensitive games will probably have to wait for quantum-entanglement to become mainstream.
On non-reaction-critical games however, this will probably happen a lot sooner, one might argue that it has already with Facebook-games and such. I expect HTML5 canvas will do wonders for that market.
I play Company Of Heroes through Steam under Wine. The suckiest part of that experience is Steam itself, which is buggy under all wine-versions I've tested. (Problems with GUI, web-browser-parts works poorly, and sometimes input-focus gets completely screwed up).
I think Steam could have lot to win by pairing up with, say Crossover, and publishing known-working games for Linux under Crossover. (And of course native Linux Games as well).
I'm not sure what sort of investment or difficulty level demos involve, but I'm gonna go out on a really short limb here and say it's a lot more difficult and expensive than slapping six words at the top of the newspaper page or splicing together some shots of the completed film or television show with a voice-over.
Doubtful. Disable a few entries in the menu, pick out a few levels, and strip any asset not referenced by those levels. (Could probably even be done by automatic garbage-collection.), and add the word "demo" to a splash-screen, and "please buy" instead of the usual end-screens. Just the professional voice-over for the trailer would probably cover all the costs of a demo.
Even for the consumer, video game reviews are generally of higher quality and more consistency than those for other products, and I'm not sure I value a demo over simple word of mouth and professional critique.
I do. Everything is relative, but my experience of the demo is only relative to me, and therefore absolute (to me).
For the professional review, I first have quite a lot of digging to do to determine my similarity to the reviewer (hates sports-games, loves deep RPG:s...), before I can asses the relevance of his/her review. For a word-of-mouth, I'm not even certain to know much about the source, it's more or less smoke. (Interesting, but not comparable to playing through a level of my own).
As a comparison, I would guess that I end up buying ~1/4 out of all demoed games I try. 14 out of my 19 (yes, I counted) PS3 BD-games are games I purchase either after trying the demo, or having played the prequel.
For the publishers, demos may have an unacceptable return on investment...
Of course, using one self as a reference is always risky, but if my case (and those around me) are ANY indication at all, I wouldn't bet on it.
I'm experiencing something similar, although, I must say it's not the "coding" that kills me, but the amount of context-switching I'm forced to do on a daily basis.
First, just let me explain my work. I'm the only developer in a technical department of 15 people, in a small local branch of a much bigger company. As such, I serve both various developing needs of my local department, as well as other departments in my branch, most support-systems for first, second and third-line support staff, as well as a multitude of network partners. Except for the development duties, there's an emphasis on last-line support, IP-networking, and product management.
What I've discovered, is that the single largest mental challenge for me, is being forced to radical context-switches, often without advanced warning, and many, many times a day. One minute, I may be working with low-level IP-protocol-debugging in hex-dumps and bit-masks, and 5-minutes later supporting the operational staff with ongoing database-issues, 15-minutes later forced into a spontaneous meeting about human-resources content-managment problems in the customer-support systems, being interupted by another scheduled meeting about conceptual architecture and product management.
All the individual context-switches is what really hurts me. I've reached a point when whenever a colleague shows up, or whenever the next bug is in a different system than the last, I almost experience physical pain, and mental pictures of a harddrive about to give up, trying to chug in those long-gone swap-pages.
You mention "maintaining a couple of dozen web apps"? Even if the technologies may or may not vary much for you, is it possible that the context and nature of the different apps are much varying, giving you a similar problem to mine? That is, forced to "switch project" often, and spending a lot of time and energy on trying to remember the relevant details for the next bug on the list?
Otherwise, when circumstances allow, I try to work from home, turn off the cellphone and shield myself off, in order to concentrate on that specific project. With all the context-switching gone I usually don't have a problem spending at least 25-30H/week doing serious "coding" (including design. development-oriented QA, test-cases, careful versioning), more for mundane typing-heavy projects, and less for more demanding designs, either due to unusual requirements, really tricky algorithms, or simply big complex stuff.
For challenging projects however, there's certainly a "burstiness" to the productivity. Some problems simply needs to be processed "offline" for a while, before they can be solved, but I think that goes for any mentally challenging activity for anyone. Most people I talk to tells me that they solve problems best while sleeping, so maybe you should find someplace to hide for a nap during the workday?;)
Hopefully not too long. Increased diversity and visible lock-ins will only provided ammunition for those of us that cares about open systems and open standards.
I've recently read the short description of the MPEG-LA license terms for broadcasters. (Not the full licenses, though)
If I understand it correctly, by purchasing a license, you're allowed to use h.264 for YOUR distribution, but the terms does not mention re-licensing to third party. To my best guess, that would mean re-licensing is not allowed.
But, and here's the catch, when YouTube-videos are embedded into other sites (Facebook, or Joe Shmoe:s blog) isn't that a form of re-sale to third party?
There is one danger with hardlinks that should not be forgotten. Hardlinks are not copy-on-write (and AFAIK, can't be made COW?), which means that if files get linked in the de-duplication-process, updates to either file will contaminate the other.
A practical example where this WOULD be a definite problem could be a double-buffered application, that for consistency always keeps a "backup" of it's config. During idle, this file could be identical to the "live" file, and hard-linking them could completely destroy the consistency feature of the app.
Another scenario would be having a file on your desktop of some family photo you want to mess around with, also in archive. Hardlink them, and editing the one on the desktop will overwrite the one in the archive. (Under some conditions, I.E. no move-operations done by the editing app)
And the fee for linux is the cost of the admins -- the people who are good in the environment know they are good and their price goes up every year -- it takes about three years to become really proficient at most MS products - it might take half a year to really understand the linux environment and methodology if you are unfamiliar and then another 4 - 5 to gain that same profiencency equivalent.
Interestingly, I work for a company where the IT-department is a pure Microsoft-shop. Only windows-hosting and almost only C#-development for internal applications. In the technology department however, we operate a bunch of production-system for our customers, running mostly CentOS Linux. Lately, I've discovered that the Linux-admin-staff often writes simple script-solutions with their left hand, with equivalent complexity to systems written by trained developer in the IT department. And even though I personally often would rather see a more structured systematic approach to some problems, when listening to the end-users they almost always perceive they've gotten BETTER support and reliability from those scripts.
Point being that, a Linux-admin MAY cost a bit more than a windows-admin, and the learning-period might certainly be a bit longer, but I see much more productivity coming out of our Linux-crew than the windows-equivalent. More services hosted and administered per admin, and ~10 times the operational availability. Also, when more complex jobs needs being done (configuring network device, someone needs help with a tricky SQL query for a report, or needs someone to mirror a huge chunk of text-files into a searchable DB for performance), they usually come to the Linux-crew than the windows IS/IT.
What I will give the person who goes the linux route is that once you are profiecient in Linux - gaining the same proficeincy in other systems is cake - basically because the they are just easier to use in the first place.
Definitely matter of oppinion. Personally, I've never found anything "easy" about windows. Sure, the very limited amount of things you can do within three button-presses is usually simple enough (interestingly the same goes for modern Linux Desktops/Simple Server Setups). However, once something breaks, or you need/desire to stride outside the comfy gui-box, just forget about it. (IMHO) For example, a standard CentOS5 server install comes with high-availability software that from commercial vendors (IBM and HP, I don't know if Microsoft can even match the fully distributed transactional storage components) START at ~100K euro. For those money, I can let one high-school self-taught Linux hacker spend 2 years in researching and fine-tuning for the JUST the entrance fees of the proprietary variant. How would you estimate my chances of getting some more use out of that admin meanwhile?
My point being that your assumption that JIT does THAT much of a difference is a bit flawed. CPython still is no match when comparing to JIT Java, of course, but the difference is roughly 30X, not 10000X.
You can run Java in a purely interpreted mode like Python does, pass the -Xint argument on the command line to the java instance. You'll notice it's 10000x slower than the optimized execution paths.
I worked for an ISP of sorts, for a student campus. Students got free internet (baked into their rent), but not everyone got white IP-addresses. Instead, clients were NAT:ed behind a smaller pool of white ip-addresses, and if you wanted your own white ip, you had to motivate it.
Almost everyone that asked got a white IP, some even got two, but the point was that for the people that didn't care, we saved over-allocating IP:s (we were subnetted and given a range of IP:s from the University).
IPv6 was of course for everyone that wanted it (opt-in through web-gui). This was 2001-2005, IPv6 were implemented around 2004.
Now don't say there aren't ways forward in IPv6-transition. (At least for the ISP)
Only if that app does not have to communicate in any way with the rest of the system. What people encouraging virtualization tends to forget is that a multi-tasking OS already have means of protection. The memory an application sees is virtual, and the access to the rest of the system often enforces a security-model.
Still, however, the user has little use for isolated applications that cannot talk to others. A modern web-browser more or less requires other apps to be of any use, such as flash, a pdf viewer, maybe access to the OS centralised authentication management (stored passwords, Kerberos SSO...), and it needs to be able to store downloaded files where other applications can open them.
Fully contained and isolated apps are great for security, but crap for the user, which is when users usually starts breaking down the security enforcements to get any work done. The key is finding an appropriate balance between usability and security, which of course varies depending on security-requirements.
On the other hand, openness encourages competition, which tends to lead to better products. Naturally all browsers won't perform as well with just like they differ performance-wise in any other aspect of HTML. But as with other aspects of HTML, there will be some innovative "Chrome"-like, performing superior to the competition, forcing it to keep up.
Above all, we may finally get rid of the virus-ridden innovation-hampering mono-culture that is Flash. Performance isn't everything.
I think the article fails to weigh in the security track-record of flash. (Which I personally see as to a large extent inherent in monopolistic closed-source mono-cultures). Performance isn't everything.
This specific error should really not be written out like that at all, but the program should be written in a way that expects that things can go wrong, and that the user don't really care WHAT went wrong, but only how/if they can fix it.
Instead of expecting your user to care about the actual error, (they REALLY don't care about the file and line where it happened) you should, 1) Design your program in a way that supports hard or soft atomic "operations". (This is a good approach for many failure-scenarios in general.) 2) Undo the atomic operation that led up to the problem and show the user a "Your computer does not have enough memory for this; Things you can try; X..Y..Z...". 3) It's definitely a good idea to combine the "OOMException"-handler with some memorable picture giving a psychological identity to the problem. This allows the user to memorize the problem with the list of possible solutions you presented. (Closing some other program, chopping the Excel document up into smaller pieces, whatever worked the last time). The next time the same error appears (no-matter what source-line) they don't really have to carefully read all the info, just try what usually works for that "bursting box thingy". 4) Then there may be a small text or icon hinting "developer information", should a user ask someone who's actually interested.
Thinking about it, perhaps one thing missing for useful error dialogs, is libraries of common error-dialogs, just as we reuse libraries for most other non-app-specific part of the application? All the common user-dialogs for OOM, Printing, Harddrive-failure, Network-failure, etc could be there? A NullPointerException could be a generic "oh, developer made a dodo"-dialog with a single button "blame him" feeding stacktraces and memdumps directly to a datamining-facility where developers can find their most common Nullpointers and similar.
Indeed. I know lots of cases where people buy software, movies or music, when all they really want is the receipt.
For software, the pirated version is often just more convenient.
For movies and music, tech-junkies buy the CD/DVD/Blueray just out of honesty, but what they really want is to have it in their mediatank. Downloading is often both faster and more convenient than ripping (and possibly transcoding).
I think there might actually even be a business opportunity in selling receipts for downloaded material. I.E. imagine you have 3d-party service, company X, entitled to sell rights for movies from some media-houses. They track pirate-sites, download and verify content they own rights to, and create checksums for the content.
The consumer then runs a small program provided by X (could be open-source program to build trust), that quickly scans through downloaded material, finds content available for licensing, and then prompts the up-until-now-pirate, for an option to pay a reasonable fee for buying a license to what he/she just downloaded. X:s servers then generate a personalized, cryptographically signed receipt for the content, (and possibly store an online-backup for safekeeping). The would-be-pirate has now gotten legit.
As an example of the revenue-opportunity; take a movie like Ninja assasin. According to torrentz.com, at this very moment, there's 100k downloads going on. Estimate a "reasonable price being 6$", and that just a single percentof all pirates could be persuaded. That's six thousand dollars just THIS VERY MOMENT. Imagine an average download of that movie lasting 1 hour, and it's 1 million dollars just this week.
Too bad I did not know this when I bought the game for PS3. Had I known, I would certainly have spent my money on something else.
I would like legislation forcing DRM:ed software and content to clearly state that in the labeling. Similar to how cigarettes now has to be labeled (although that in itself is a frigthening sign of stupidity), DRM:ed software and music should come with a "this is crippled, and may stop working, or not work at all"-warning.
Let's see how well DRM would do with an informed consumer-base.
I work for a large TV Broadcast Company operating over both Cable and IPTV.
A few years ago, we did extensive testing of the products of Quiqom (now Dican) with IPTV as payload. After a few initial hickups that were fixed in firmware, we eventually were able to stream 8 TV-channels, with FTP:ing and Torrents as background-noise, and with zero packet-loss or bit-errors. (And marginal delays, even though I don't recollect specifics). I were amazed to see how well it worked. At the end, I could not come up with any reasonable load to break the thing. (Wild torrenting, heavy load, peeling of parts of the isolation of the cable...)
I'm not sure if their products can be bought outside Sweden, let alone by home users (they're really for Network Operators), but if their MoCa-based product works that great, I'd say there's a fair chance others may as well.
Yet another angle on it, is that all IE combined has been on a steady decline for a good while, now also in January.
Now for the FIRST TIME, w3counter puts IE below the 50%-line, which means that slightly over half of all users now actually DO run a more sensible browser.
In my mind, that's a sign of a fantastic, and unexpected awareness amongst computer users.
Slashdot Mirror
The real danger is actually in the grandparent.
Likely, your neighbors won't be thinking twice about security INSIDE the network, since they "know" they are secure now, making it easier to browse their pr0n collection.
It's not exactly uncommon for even moderately security-aware people to have all kinds of open doors locally behind their lovely little NAT-box, making it very relatively easy for a malicious neighbour to cause all kinds of damage.
What are the chances you pissed some neighbour off in the past month?
Q: How are you going to steal my bytes when I don't pub my SSID?
A: iwlist still shows it.
Uhm, is it possible you haven't realized by now, that "it" in the original answer does not refer to "my SSID", but to the network, as "my bytes"? I had no problems getting the point the first time, and I've seen no backpedaling.
Lovely "discussion", both of you managing to convey nothing in so many words. That's what I love about slashdot, keep the good work up!
That may take a while, at least for reaction-heavy games.
Ignoring latencies in active repeaters (routers, switches...) just the speed of light in fiber has a measurable roundtrip (about 12ms from Stockholm to Paris). Then add audio-video-encoding, which realistically needs to buffer at least a few frames to get decent compression (x264 for live-broadcast is usually varying from 500-1000ms), so that adds another 20-30ms. Then you'd either need strong FEC, or very fast re-transmission to not occasionally suffer heavy distortions in the picture (oops, there was a grenade), which either adds extra bandwidth and a little extra delay, or more delay with less bandwidth (could possibly be migigated with a failure-resilient wavelet codec). In the end, somewhere around 50ms latency is probably unavoidable, at which point all reaction-heavy games is a nono for many gamers. (Equal to ~20FPS)
To compete with graphical quality of a local game, you'd either need closer 100mbit/user to get even decent reaction-times using current and coming codecs, or a lot more for great reaction-times. That would of course be feasible for smaller edge-segments of the network, such as perhaps a gaming-server in the basement of every apartment-building, but that quickly becomes a logistics and service-nightmare.
Other than that, reaction-sensitive games will probably have to wait for quantum-entanglement to become mainstream.
On non-reaction-critical games however, this will probably happen a lot sooner, one might argue that it has already with Facebook-games and such. I expect HTML5 canvas will do wonders for that market.
I play Company Of Heroes through Steam under Wine. The suckiest part of that experience is Steam itself, which is buggy under all wine-versions I've tested. (Problems with GUI, web-browser-parts works poorly, and sometimes input-focus gets completely screwed up).
I think Steam could have lot to win by pairing up with, say Crossover, and publishing known-working games for Linux under Crossover. (And of course native Linux Games as well).
I'm not sure what sort of investment or difficulty level demos involve, but I'm gonna go out on a really short limb here and say it's a lot more difficult and expensive than slapping six words at the top of the newspaper page or splicing together some shots of the completed film or television show with a voice-over.
Doubtful. Disable a few entries in the menu, pick out a few levels, and strip any asset not referenced by those levels. (Could probably even be done by automatic garbage-collection.), and add the word "demo" to a splash-screen, and "please buy" instead of the usual end-screens. Just the professional voice-over for the trailer would probably cover all the costs of a demo.
Even for the consumer, video game reviews are generally of higher quality and more consistency than those for other products, and I'm not sure I value a demo over simple word of mouth and professional critique.
I do. Everything is relative, but my experience of the demo is only relative to me, and therefore absolute (to me).
For the professional review, I first have quite a lot of digging to do to determine my similarity to the reviewer (hates sports-games, loves deep RPG:s...), before I can asses the relevance of his/her review. For a word-of-mouth, I'm not even certain to know much about the source, it's more or less smoke. (Interesting, but not comparable to playing through a level of my own).
As a comparison, I would guess that I end up buying ~1/4 out of all demoed games I try. 14 out of my 19 (yes, I counted) PS3 BD-games are games I purchase either after trying the demo, or having played the prequel.
For the publishers, demos may have an unacceptable return on investment...
Of course, using one self as a reference is always risky, but if my case (and those around me) are ANY indication at all, I wouldn't bet on it.
I'm experiencing something similar, although, I must say it's not the "coding" that kills me, but the amount of context-switching I'm forced to do on a daily basis.
First, just let me explain my work. I'm the only developer in a technical department of 15 people, in a small local branch of a much bigger company. As such, I serve both various developing needs of my local department, as well as other departments in my branch, most support-systems for first, second and third-line support staff, as well as a multitude of network partners. Except for the development duties, there's an emphasis on last-line support, IP-networking, and product management.
What I've discovered, is that the single largest mental challenge for me, is being forced to radical context-switches, often without advanced warning, and many, many times a day. One minute, I may be working with low-level IP-protocol-debugging in hex-dumps and bit-masks, and 5-minutes later supporting the operational staff with ongoing database-issues, 15-minutes later forced into a spontaneous meeting about human-resources content-managment problems in the customer-support systems, being interupted by another scheduled meeting about conceptual architecture and product management.
All the individual context-switches is what really hurts me. I've reached a point when whenever a colleague shows up, or whenever the next bug is in a different system than the last, I almost experience physical pain, and mental pictures of a harddrive about to give up, trying to chug in those long-gone swap-pages.
You mention "maintaining a couple of dozen web apps"? Even if the technologies may or may not vary much for you, is it possible that the context and nature of the different apps are much varying, giving you a similar problem to mine? That is, forced to "switch project" often, and spending a lot of time and energy on trying to remember the relevant details for the next bug on the list?
Otherwise, when circumstances allow, I try to work from home, turn off the cellphone and shield myself off, in order to concentrate on that specific project. With all the context-switching gone I usually don't have a problem spending at least 25-30H/week doing serious "coding" (including design. development-oriented QA, test-cases, careful versioning), more for mundane typing-heavy projects, and less for more demanding designs, either due to unusual requirements, really tricky algorithms, or simply big complex stuff.
For challenging projects however, there's certainly a "burstiness" to the productivity. Some problems simply needs to be processed "offline" for a while, before they can be solved, but I think that goes for any mentally challenging activity for anyone. Most people I talk to tells me that they solve problems best while sleeping, so maybe you should find someplace to hide for a nap during the workday? ;)
Definitely, and I'm pretty sure the apps there could be of highest standards, well integrated with everything else.
How easy it would be to attract developers to that kind of lock-in, is another question.
Hopefully not too long. Increased diversity and visible lock-ins will only provided ammunition for those of us that cares about open systems and open standards.
I've recently read the short description of the MPEG-LA license terms for broadcasters. (Not the full licenses, though)
If I understand it correctly, by purchasing a license, you're allowed to use h.264 for YOUR distribution, but the terms does not mention re-licensing to third party. To my best guess, that would mean re-licensing is not allowed.
But, and here's the catch, when YouTube-videos are embedded into other sites (Facebook, or Joe Shmoe:s blog) isn't that a form of re-sale to third party?
Can someone with more insight comment on this?
Still, whenever a Tab hangs in my Chromium, usually most, or all other tab dies as well, occasionally entire chromium.
There is one danger with hardlinks that should not be forgotten. Hardlinks are not copy-on-write (and AFAIK, can't be made COW?), which means that if files get linked in the de-duplication-process, updates to either file will contaminate the other.
A practical example where this WOULD be a definite problem could be a double-buffered application, that for consistency always keeps a "backup" of it's config. During idle, this file could be identical to the "live" file, and hard-linking them could completely destroy the consistency feature of the app.
Another scenario would be having a file on your desktop of some family photo you want to mess around with, also in archive. Hardlink them, and editing the one on the desktop will overwrite the one in the archive. (Under some conditions, I.E. no move-operations done by the editing app)
And the fee for linux is the cost of the admins -- the people who are good in the environment know they are good and their price goes up every year -- it takes about three years to become really proficient at most MS products - it might take half a year to really understand the linux environment and methodology if you are unfamiliar and then another 4 - 5 to gain that same profiencency equivalent.
Interestingly, I work for a company where the IT-department is a pure Microsoft-shop. Only windows-hosting and almost only C#-development for internal applications. In the technology department however, we operate a bunch of production-system for our customers, running mostly CentOS Linux. Lately, I've discovered that the Linux-admin-staff often writes simple script-solutions with their left hand, with equivalent complexity to systems written by trained developer in the IT department. And even though I personally often would rather see a more structured systematic approach to some problems, when listening to the end-users they almost always perceive they've gotten BETTER support and reliability from those scripts.
Point being that, a Linux-admin MAY cost a bit more than a windows-admin, and the learning-period might certainly be a bit longer, but I see much more productivity coming out of our Linux-crew than the windows-equivalent. More services hosted and administered per admin, and ~10 times the operational availability. Also, when more complex jobs needs being done (configuring network device, someone needs help with a tricky SQL query for a report, or needs someone to mirror a huge chunk of text-files into a searchable DB for performance), they usually come to the Linux-crew than the windows IS/IT.
What I will give the person who goes the linux route is that once you are profiecient in Linux - gaining the same proficeincy in other systems is cake - basically because the they are just easier to use in the first place.
Definitely matter of oppinion. Personally, I've never found anything "easy" about windows. Sure, the very limited amount of things you can do within three button-presses is usually simple enough (interestingly the same goes for modern Linux Desktops/Simple Server Setups). However, once something breaks, or you need/desire to stride outside the comfy gui-box, just forget about it. (IMHO) For example, a standard CentOS5 server install comes with high-availability software that from commercial vendors (IBM and HP, I don't know if Microsoft can even match the fully distributed transactional storage components) START at ~100K euro. For those money, I can let one high-school self-taught Linux hacker spend 2 years in researching and fine-tuning for the JUST the entrance fees of the proprietary variant. How would you estimate my chances of getting some more use out of that admin meanwhile?
Or at least link-posting. The link should of course be http://shootout.alioth.debian.org/u64/benchmark.php?test=all&lang=python3&lang2=java
The question still remains, 10000X seems like a grave exaggeration. (Which is confirmed by http://shootout.alioth.debian.org/u64/benchmark.php?test=all&lang=javaxint&lang2=java)
My point being that your assumption that JIT does THAT much of a difference is a bit flawed. CPython still is no match when comparing to JIT Java, of course, but the difference is roughly 30X, not 10000X.
You can run Java in a purely interpreted mode like Python does, pass the -Xint argument on the command line to the java instance. You'll notice it's 10000x slower than the optimized execution paths.
Then why isn't Python 10000x times slower than Java? http://shootout.alioth.debian.org/u64/benchmark.php?test=all&lang=python3&lang2=javaxint
I worked for an ISP of sorts, for a student campus. Students got free internet (baked into their rent), but not everyone got white IP-addresses. Instead, clients were NAT:ed behind a smaller pool of white ip-addresses, and if you wanted your own white ip, you had to motivate it.
Almost everyone that asked got a white IP, some even got two, but the point was that for the people that didn't care, we saved over-allocating IP:s (we were subnetted and given a range of IP:s from the University).
IPv6 was of course for everyone that wanted it (opt-in through web-gui). This was 2001-2005, IPv6 were implemented around 2004.
Now don't say there aren't ways forward in IPv6-transition. (At least for the ISP)
Only if that app does not have to communicate in any way with the rest of the system. What people encouraging virtualization tends to forget is that a multi-tasking OS already have means of protection. The memory an application sees is virtual, and the access to the rest of the system often enforces a security-model.
Still, however, the user has little use for isolated applications that cannot talk to others. A modern web-browser more or less requires other apps to be of any use, such as flash, a pdf viewer, maybe access to the OS centralised authentication management (stored passwords, Kerberos SSO...), and it needs to be able to store downloaded files where other applications can open them.
Fully contained and isolated apps are great for security, but crap for the user, which is when users usually starts breaking down the security enforcements to get any work done. The key is finding an appropriate balance between usability and security, which of course varies depending on security-requirements.
On the other hand, openness encourages competition, which tends to lead to better products. Naturally all browsers won't perform as well with just like they differ performance-wise in any other aspect of HTML. But as with other aspects of HTML, there will be some innovative "Chrome"-like, performing superior to the competition, forcing it to keep up.
Above all, we may finally get rid of the virus-ridden innovation-hampering mono-culture that is Flash. Performance isn't everything.
+1.
I think the article fails to weigh in the security track-record of flash. (Which I personally see as to a large extent inherent in monopolistic closed-source mono-cultures). Performance isn't everything.
This is a great example of the problem.
This specific error should really not be written out like that at all, but the program should be written in a way that expects that things can go wrong, and that the user don't really care WHAT went wrong, but only how/if they can fix it.
Instead of expecting your user to care about the actual error, (they REALLY don't care about the file and line where it happened) you should,
1) Design your program in a way that supports hard or soft atomic "operations". (This is a good approach for many failure-scenarios in general.)
2) Undo the atomic operation that led up to the problem and show the user a "Your computer does not have enough memory for this; Things you can try; X..Y..Z...".
3) It's definitely a good idea to combine the "OOMException"-handler with some memorable picture giving a psychological identity to the problem. This allows the user to memorize the problem with the list of possible solutions you presented. (Closing some other program, chopping the Excel document up into smaller pieces, whatever worked the last time). The next time the same error appears (no-matter what source-line) they don't really have to carefully read all the info, just try what usually works for that "bursting box thingy".
4) Then there may be a small text or icon hinting "developer information", should a user ask someone who's actually interested.
Thinking about it, perhaps one thing missing for useful error dialogs, is libraries of common error-dialogs, just as we reuse libraries for most other non-app-specific part of the application? All the common user-dialogs for OOM, Printing, Harddrive-failure, Network-failure, etc could be there? A NullPointerException could be a generic "oh, developer made a dodo"-dialog with a single button "blame him" feeding stacktraces and memdumps directly to a datamining-facility where developers can find their most common Nullpointers and similar.
Indeed. I know lots of cases where people buy software, movies or music, when all they really want is the receipt.
For software, the pirated version is often just more convenient.
For movies and music, tech-junkies buy the CD/DVD/Blueray just out of honesty, but what they really want is to have it in their mediatank. Downloading is often both faster and more convenient than ripping (and possibly transcoding).
I think there might actually even be a business opportunity in selling receipts for downloaded material. I.E. imagine you have 3d-party service, company X, entitled to sell rights for movies from some media-houses. They track pirate-sites, download and verify content they own rights to, and create checksums for the content.
The consumer then runs a small program provided by X (could be open-source program to build trust), that quickly scans through downloaded material, finds content available for licensing, and then prompts the up-until-now-pirate, for an option to pay a reasonable fee for buying a license to what he/she just downloaded. X:s servers then generate a personalized, cryptographically signed receipt for the content, (and possibly store an online-backup for safekeeping). The would-be-pirate has now gotten legit.
As an example of the revenue-opportunity; take a movie like Ninja assasin. According to torrentz.com, at this very moment, there's 100k downloads going on. Estimate a "reasonable price being 6$", and that just a single percentof all pirates could be persuaded. That's six thousand dollars just THIS VERY MOMENT. Imagine an average download of that movie lasting 1 hour, and it's 1 million dollars just this week.
Too bad I did not know this when I bought the game for PS3. Had I known, I would certainly have spent my money on something else.
I would like legislation forcing DRM:ed software and content to clearly state that in the labeling. Similar to how cigarettes now has to be labeled (although that in itself is a frigthening sign of stupidity), DRM:ed software and music should come with a "this is crippled, and may stop working, or not work at all"-warning.
Let's see how well DRM would do with an informed consumer-base.
There are also a bunch of programs that is not so clear about them being P2P. Spotify and Voddler comes to mind, but there are more.
I agree with parent.
I work for a large TV Broadcast Company operating over both Cable and IPTV.
A few years ago, we did extensive testing of the products of Quiqom (now Dican) with IPTV as payload. After a few initial hickups that were fixed in firmware, we eventually were able to stream 8 TV-channels, with FTP:ing and Torrents as background-noise, and with zero packet-loss or bit-errors. (And marginal delays, even though I don't recollect specifics). I were amazed to see how well it worked. At the end, I could not come up with any reasonable load to break the thing. (Wild torrenting, heavy load, peeling of parts of the isolation of the cable...)
I'm not sure if their products can be bought outside Sweden, let alone by home users (they're really for Network Operators), but if their MoCa-based product works that great, I'd say there's a fair chance others may as well.
Probably. Although, pretty much all browser-analytics show the same trend. They just disagree on the actual numbers. :)
Yet another angle on it, is that all IE combined has been on a steady decline for a good while, now also in January.
Now for the FIRST TIME, w3counter puts IE below the 50%-line, which means that slightly over half of all users now actually DO run a more sensible browser.
In my mind, that's a sign of a fantastic, and unexpected awareness amongst computer users.