There's a bunch of things firefox has included in the core browser that probably should be replaced with an addon. There have been many examples of this already. What they are missing here is a prompt on upgrade if their change in behaviour applies to you so you aren't surprised by it.
Most users don't read the release notes, or follow some obscure blog where this change may have been discussed.
How many users will hit the "block all cookies from this domain" button, and then blame firefox for being broken? There are good reasons for reducing the number of ways users can screw up their configuration.
Right, Firefox have been simplifying options that break web sites for years. If you want to mess with your browsing experience in a way that might break how websites work, install an addon.
There's no point having what amounts to a "Make all websites work" checkbox.
An early comment to a story may seem correct and get modded up. Only to be proven false and then modded down. This is fine, it should be possible to make it disappear.
Games usually close without prompting when you press ALT+F4. They can handle the WM_CLOSE & WM_QUERYENDSESSION events; open a "Do you really want to quit" dialog and prevent a logoff. But I haven't seen many games that do this properly.
I'm sure you don't want to funnel traffic to some other middleman who is only trying to boost his own advertising revenue. As much as possible, find the original source of a story and link to that prominently. Science reporting? Direct link to arXiv or similar. We're going to want to dissect the paper in the comments anyway. If there are other sources with more context or analysis, fine keep them too.
Good. That deserves to be on the front page, at least once you've got a solid plan for how you are going to fix it.
Github is the obvious competitor you have to beat, or at least distinguish yourself from. If you're going to focus on source control you have to beat github on features. Make it easy for people to contribute to projects, while also respecting the contribution rules of that project.
My suggestion would be to focus on the things that github doesn't do. If you want to be a trusted place for downloading software, help us to trust the integrity of the binaries you are hosting. Host build & test servers that can be used to produce repeatable builds.
Code review before the "RockStar" can merge his code in, should give the same result. Your new feature is not done until someone else can use it & maintain it. Otherwise the business is screwed if you get hit by a bus.
Starting a policy of reviewing all code may seem like a waste of time. But as a team works together more, code reviews should get faster.
To cover some of TFA's "Downsides";
Working on stories which have dependencies is hard
Yes. If you find you need to reuse some common code between two feature branches, first refactor & rebase that code into a separate feature branch. If possible, merge that code to master first. Then you can both merge it cleanly. A branch is not a "feature" or a "use case", it's a patch series that you can merge without breaking production. You don't have to wait until a whole "story" (or whatever you call it) is complete.
Reviewing big changes is hard
Try to split big changes into a series of smaller commits before pushing for review. Again, this may involve rebasing & reworking patches that you have already written. Don't get too attached to the commit's you have already created. I don't want to review all of your experiments while you were working out what not to do.
Out of 74,207,281(-ish) tested values for n, this is only the 49th prime found. If you tested the first 74,207,281 odd numbers you would have found more that 5 million primes.
... may be the AMD GPU itself is simply not powerful enough to use that bandwidth effectively...
Building the right balance of compute units, memory capacity & bandwidth is a hard problem. Games developers will target high frame-rates on their target hardware, optimising or cutting back features until everything works well enough.
We will have to wait and see if developers will find creative ways to use this bandwidth increase.
There is NO technical means by which you can have a backdoor which is only usable by one government
Counter argument; Dual_EC_DRBG. Only an entity that knows the hidden key that the algorithm is using can deduce the internal state of this RNG.
Backdoors are a problem, and I'm not suggesting that they should be implemented. But it is possible to build one so that only the person with the right secret key can open it.
If you think it's only a matter of faster technology, or better maths. Here's another impossible problem for you; Write down the decimal expansion of Graham's Number.
There are hard limits to what we can compute with the matter, space and time available to us within this universe.
It's the top end tier on a kickstarter, so of course it's a waste of time and money. The guts are built around an Arm & FPGA. This is a machines designed by a hardware hacker, for a hardware hacker. It's not for everyone.
Cracking Dual_EC requires knowledge of a secret that was used to generate the elliptic curve parameters it uses. The NSA published a set of parameters as part of the proposed standard. If these are the parameters that Juniper used, then only the NSA can deduce the internal state of the random number generator.
There's no point to anyone else adding this backdoor, unless they are friends with the NSA.
The recession was inevitable, and we still haven't dealt with all of the fallout. Sub prime housing loans were the straw that broke the camels back, but even that is only a symptom of a much larger underlying problem.
Economists don't understand the economy. They are experts at trying to explain why banks, debt and money don't matter. When in the real world, they matter a great deal.
Until economists study and understand the actual role of bank debt, and act to constrain lending to constructive activities, we will continue to make the same mistakes.
the proportion of readers using ad blockers dropped from 23% to the single digits when faced with the choice to turn off the software or pay... Over two-thirds of the users concerned switched off their adblocker.
Did they? Or did they simply not come back?
Of course with the developer tools built into browsers these days, it only takes a few clicks to delete the nag layer and get to the underlying content. I wonder how they count me in their statistics?
That only works if your code is not stored as text. If your IDE parsed the code, and re-rendered it with your preferences, this might be bearable. Otherwise there's no way to work together as a team.
Jay Radcliffe gave a talk at Black Hat, showing how his personal insulin pump could be hacked. If he wants to know that the security research he is planning will not run afoul of the DMCA, he's going to need an army of lawyers to comb through the DMCA rulemaking performed every three years by the Librarian of Congress. This process is a useless garbage train that’s gone completely off the tracks. Copyright law is rarely sensible, but at this point, DMCA section 1201 has spiraled entirely out of the realm of copyright and into a Kafka-esque hellscape.
Nigerian scams, certainly. But lots of spam is attempting to get you to buy something online, or compromise your PC in some way. Plus of course, there's the problem of spoofed from addresses causing unintended damage to 3rd parties.
Slashdot Mirror
There's a bunch of things firefox has included in the core browser that probably should be replaced with an addon. There have been many examples of this already. What they are missing here is a prompt on upgrade if their change in behaviour applies to you so you aren't surprised by it.
Most users don't read the release notes, or follow some obscure blog where this change may have been discussed.
How many users will hit the "block all cookies from this domain" button, and then blame firefox for being broken? There are good reasons for reducing the number of ways users can screw up their configuration.
Right, Firefox have been simplifying options that break web sites for years. If you want to mess with your browsing experience in a way that might break how websites work, install an addon.
There's no point having what amounts to a "Make all websites work" checkbox.
An early comment to a story may seem correct and get modded up. Only to be proven false and then modded down. This is fine, it should be possible to make it disappear.
Games usually close without prompting when you press ALT+F4. They can handle the WM_CLOSE & WM_QUERYENDSESSION events; open a "Do you really want to quit" dialog and prevent a logoff. But I haven't seen many games that do this properly.
Bingo, they have a list of existing software that they recommend.
I'm sure you don't want to funnel traffic to some other middleman who is only trying to boost his own advertising revenue. As much as possible, find the original source of a story and link to that prominently. Science reporting? Direct link to arXiv or similar. We're going to want to dissect the paper in the comments anyway. If there are other sources with more context or analysis, fine keep them too.
Good. That deserves to be on the front page, at least once you've got a solid plan for how you are going to fix it.
Github is the obvious competitor you have to beat, or at least distinguish yourself from. If you're going to focus on source control you have to beat github on features. Make it easy for people to contribute to projects, while also respecting the contribution rules of that project.
My suggestion would be to focus on the things that github doesn't do. If you want to be a trusted place for downloading software, help us to trust the integrity of the binaries you are hosting. Host build & test servers that can be used to produce repeatable builds.
Code review before the "RockStar" can merge his code in, should give the same result. Your new feature is not done until someone else can use it & maintain it. Otherwise the business is screwed if you get hit by a bus.
Starting a policy of reviewing all code may seem like a waste of time. But as a team works together more, code reviews should get faster.
To cover some of TFA's "Downsides";
Working on stories which have dependencies is hard
Yes. If you find you need to reuse some common code between two feature branches, first refactor & rebase that code into a separate feature branch. If possible, merge that code to master first. Then you can both merge it cleanly. A branch is not a "feature" or a "use case", it's a patch series that you can merge without breaking production. You don't have to wait until a whole "story" (or whatever you call it) is complete.
Reviewing big changes is hard
Try to split big changes into a series of smaller commits before pushing for review. Again, this may involve rebasing & reworking patches that you have already written. Don't get too attached to the commit's you have already created. I don't want to review all of your experiments while you were working out what not to do.
No, absolutely not.
Out of 74,207,281(-ish) tested values for n, this is only the 49th prime found. If you tested the first 74,207,281 odd numbers you would have found more that 5 million primes.
... may be the AMD GPU itself is simply not powerful enough to use that bandwidth effectively ...
Building the right balance of compute units, memory capacity & bandwidth is a hard problem. Games developers will target high frame-rates on their target hardware, optimising or cutting back features until everything works well enough.
We will have to wait and see if developers will find creative ways to use this bandwidth increase.
There is NO technical means by which you can have a backdoor which is only usable by one government
Counter argument; Dual_EC_DRBG. Only an entity that knows the hidden key that the algorithm is using can deduce the internal state of this RNG.
Backdoors are a problem, and I'm not suggesting that they should be implemented. But it is possible to build one so that only the person with the right secret key can open it.
If you think it's only a matter of faster technology, or better maths. Here's another impossible problem for you; Write down the decimal expansion of Graham's Number.
There are hard limits to what we can compute with the matter, space and time available to us within this universe.
It's the top end tier on a kickstarter, so of course it's a waste of time and money. The guts are built around an Arm & FPGA. This is a machines designed by a hardware hacker, for a hardware hacker. It's not for everyone.
Cracking Dual_EC requires knowledge of a secret that was used to generate the elliptic curve parameters it uses. The NSA published a set of parameters as part of the proposed standard. If these are the parameters that Juniper used, then only the NSA can deduce the internal state of the random number generator.
There's no point to anyone else adding this backdoor, unless they are friends with the NSA.
Opening the link in a private window seems to work in firefox / iceweasel.
Or a FedEx plane full of MicroSD cards.
The recession was inevitable, and we still haven't dealt with all of the fallout. Sub prime housing loans were the straw that broke the camels back, but even that is only a symptom of a much larger underlying problem.
Economists don't understand the economy. They are experts at trying to explain why banks, debt and money don't matter. When in the real world, they matter a great deal.
Until economists study and understand the actual role of bank debt, and act to constrain lending to constructive activities, we will continue to make the same mistakes.
Chrome's PNaCl defines a stable sub-set of LLVM IR, targeting a generic little-endian 32-bit CPU.
That's why he included thisAlgorithmBecomingSkynetCost = 999999999.
the proportion of readers using ad blockers dropped from 23% to the single digits when faced with the choice to turn off the software or pay ... Over two-thirds of the users concerned switched off their adblocker.
Did they? Or did they simply not come back?
Of course with the developer tools built into browsers these days, it only takes a few clicks to delete the nag layer and get to the underlying content. I wonder how they count me in their statistics?
Or for Weird Al fans; mission statement
That only works if your code is not stored as text. If your IDE parsed the code, and re-rendered it with your preferences, this might be bearable. Otherwise there's no way to work together as a team.
Is this more to your liking?
Jay Radcliffe gave a talk at Black Hat, showing how his personal insulin pump could be hacked. If he wants to know that the security research he is planning will not run afoul of the DMCA, he's going to need an army of lawyers to comb through the DMCA rulemaking performed every three years by the Librarian of Congress. This process is a useless garbage train that’s gone completely off the tracks. Copyright law is rarely sensible, but at this point, DMCA section 1201 has spiraled entirely out of the realm of copyright and into a Kafka-esque hellscape.
Nigerian scams, certainly. But lots of spam is attempting to get you to buy something online, or compromise your PC in some way. Plus of course, there's the problem of spoofed from addresses causing unintended damage to 3rd parties.