Serious question: could you explain why SPDY is retarded? I'm actually curious what's wrong with, not knowing much about it other than having read some material when it first came out..
FTC has a do not call list that robocallers are obliged to wash their lists against. Not saying they all do it, but I put my numbers on that list and I get very few automated or human cold calls any more (and none from groups I haven't interacted with before).
That's very fair - good points. My thinking is that if AT&T can get me 3mbps DSL over copper for $30/month, the physical plant costs can't be the driver in pricing (and I'd think physical plant costs are probably lower for cable since the infrastructure is way less screwed up than copper pairs that the phone cos deal with)? Granted the phone co infrastructure is fully paid for so maybe that's an issue?
Of course DSL is less mbps but check out comcast pricing: you pay $60/month for 16mbps on cable. If you don't want 16mbps you can get 6mbps $50/month or 1.5mbps for $27/mo. The discontinuity in pricing could be related to loss leaders or physical plant (invariant) costs as you suggest, but given how cheap they are willing to bundle two or three services together, it smells like they are doing everything they can to encourage you to keep your cable subscription and not try to cut the cord and go internet only.
That makes sense if Comcast makes more money on cable subscribers than on internet subs? And why would they buy NBC if they weren't worried about cable subs cutting the cord? If they aren't in the content business then when their fat revenue from cable dries up, they become just another earthlink - big whoop.
I'm not saying I have the answers but the way they are running their business seems to suggest this line of thinking (and folks inside the FCC took this as common knowledge when I was there -- not that they necessarily have the inside track).. Curious what you think..
All votes today are not hand counted visually by a human (at least in California where I'm from), so I'm not sure what you're talking about, or rather I think this train has already left the station. Today, I physically mark a paper ballot, but the count is done by an optical scanner. Detection of fraud in that scanner can be done by random sampling, but I don't know for sure when or how this is actually done in CA.
I do know that random sampling as a practice can detect significant deviations in voter intention and recorded counts, and is used widely in ensuring fair voting in many countries around the world.
I agree with your sentiment that we shouldn't fool around with this system like we do with health records, banking or other so called "critical" services. This one critical service is far far more vulnerable and important than all the others..
The thing that chaps me is that I have to pay $60/month for internet from a cable company (and the phone company can't for whatever reason get optical or copper internet of any kind to me -- I live in downtown DC). For $70/month I could get basic cable and internet, but I don't want it. Since basic cable alone costs $40, it's clear that this pricing is b/c cable co's don't want internet service to be cheap, so they are artificially holding the price high, to retain cable subscribers, who are far, far more profitable than their internet customers (even when either customer is paying the same dollar value).
Sorry - couldn't resist. There are some channels which pay to get access to viewers, whereas all the stations you want to watch are the other way around. If a station can make money by providing old movies which cost very little to license, then they are likely to make themselves free to cable, or even pay for a station. If content is expensive and/or if consumer demand is very high for something, then cable will pay to get it on their roster so as to have a competitive product.. Ugly business.
This is true for quantity as well. If I watch very little, I am subsidizing those who watch a lot with the all you can eat cable models. Netflix and Hulu are basically the same models - Netflix $8/month all the movies you can watch. Amazon movie rentals reward less viewing - you rent by title.. So this subsidy factor goes in a lot of directions.
Not sure if the analogy holds, but with internet service, Australia has had pay per byte models which some analysts argued (inside the FCC when I worked there) that this retarded AU's investments in internet infrastructure. US isn't that great for infrastructure either so I'm not sure I buy the argument, but the idea is that "all you can eat" is pro-innovation and development, and "pay as you go" offers more consumer choices, and lower average costs (at the cost of longer term stagnation in capital formation and investment in new services).
You're joking right? If not, the random sample is to check that all the votes that were counted by the machine are reflective of what the voters actually attempted to vote for (detecting mismatch between voter intention and what the machine is producing for counting). All votes are counted in any of these systems.
I don't fold my current voting sheet. I carry it from the booth to the reader wide out in the open where anyone can see it. If I want to stuff it in my shirt, or hold it against my chest I guess I could, but even then I have to mark both sides of the ballot so the under-the-shirt method is the only way I guess.
Anyway, this problem exists today at least in my county with regular paper ballots.
Yeah - really. The point as I understand UEFI is so that it can compete with desktop boxes and other baked in hardware where there is a relatively strong assurance that the stack of software that is currently running is "known good" from the operating system's perspective. This allows you to provide streaming services which are harder to rip and things like that. This in turn will probably make more movie and content companies make their stuff available on PC's.
This is just "trusted computing" if you remember that whole thing all over again. It's not about preventing you from running Linux, it's about preventing you from getting streaming movies on Linux, b/c it's harder to guarantee (to the movie's owner that) the stream is uncrackable..
Given that TC didn't work out so well (cracks on xbox360 and PS3) this may not play out the way they intend, but who knows maybe they'll finally get their dream of locking out people from these software stacks..
So companies who create/sell GPL aren't trying to benefit from freeriding on GPL? Like SugarCRM, Compiere or "old MySQL?" My experience is that (some) GPL software is released by companies who want to specifically prevent other companies from horning in on their business by re-using their codebase in a competitive way (no one but the original rights holder can release proprietary extensions or resell the product under other licenses). Not to start and GPL/academic license debate - but can you point to some companies that release academic license OSS who have business models like SugarCRM or the "old MySQL?" What am I missing?
Read this article to see how business-people see GPL: http://www.crmoutsiders.com/2010/07/15/some-thoughts-on-open-from-sugarcrm-ceo-larry-augustin/ -- GPL gives them a business model to keep others from selling the same thing that they sell to their customers (the core GPL version plus proprietary extensions).. Everyone else could only sell core GPL plus GPL extensions, and of course Sugar could reincorporate those gpl extensions into its core eliminating the competitor's differentiating features..
Wait - I recall reading a while back that Google isn't releasing 3.0-3.1 as open source but that they have said they *intend* to release future revs 3.3+ (or something like that) again as open? Am I misremembering? Is this a blip in the scheme of things or is Google in fact never planning to release another OSS version of Android? Anyone?
While I concede that Google's android app store is crappier than Apple's, it's possible that Google's goal with Android is very different from Apple's with the iPhone. Google may be trying to prevent someone like Apple or Microsoft from closing out the phone market in the way Microsoft did on the PC market years ago. If Google prevents that, then they will be able to continue to deliver search and ads on every platform.. I think this is probably a big reason why they are invested so heavily in Android: not because it is a profit center, but it is a strategic market force to keep their main lines of profit solid as the industry moves to tablets and phones from PCs and laptops..
In which case a somewhat crappier App environment while not desirable isn't that bad either for them.. So long as users are still adopting it and preventing a market unification around the iphone, which looked inevitable (at least to me) just a few years ago.
Love it - great post. I'm on this boat too. If your shit is so simple you just want a pile of data in a real fast container, use couch or mongo. If you want SQL b/c you need SQL use Pg. That is all.
Interesting. Possibly this is a case that demonstrates why BSD/MIT/Apache licenses are superior to GPL? When Oracle acquires a copyright title on a product such as MySQL released under GPL license, they can release closed source mods to the code and convert the community to a locked-in position, whereas no other player can do the same? Everyone but Oracle is "stuck" with a viral license now and so Oracle has a huge advantage in terms of controlling the community with proprietary extensions? Hmm.
I used Ruby/Rails with Postgres just fine on a very complex project, and I think that's a reasonably agile environment? The database migrations functions in Rails permitted management of state of database in dev, test, staging and production perfectly. It was actually the best setup I've ever had for that. You could see exactly what changes happened on each system, and you could very easily roll your code back to a particular date, roll the database back to the same date, and the structures and system data, lookups, etc would all get updated to reflect their state at that time. It made it so much easier to replicate a bug that might be happening in production, when your dev box was far in advance -- just check out the correct version, run your backward migration and you're testing on a system that's the same as production - no need to even create a new database or anything.
Postgres worked very well for all this. I think what you're saying is that *the way you like to manage these issues* doesn't work in Postgres, and that may very well be true. But that's not to say that you can't easily manage those issues in Postgres.
The converse is also true - I know a few people who have iphone3 phones and who upgraded to iphone4 b/c the regular upgrades made their phones unusable. I suggested they jail break the iphone3 and go back to an earlier OS version (not even sure if this is possible not being an iphone user) but they weren't interested.
My Droid Eris is a lot slower than it used to be due to all the upgrades, but of course I do know that I can back-grade to an earlier OS version to get some speed back if it becomes unworkable for me.
But the carriers and manufacturers have plenty of incentive to over-upgrade your existing device to encourage you to go buy a new one. I wouldn't be surprised to start seeing a final "upgrade" release timed just about 2 years after the initial release so as to encourage users with their now too slow phones to go get new ones..
Great post - thanks. I'll review this stuff.. One thought on the 100x overpop comment -- you might be interested to read the book 1491 which discusses some reasonably mainstream archaeology theories that suggest north america may have supported populations approaching 100M prior to contact. And that very early contact spread disease through the Americas as a plague worse than the black death in Europe (33% population loss during that). So when Europeans first start actually walking around and looking for civilizations they see ragtag remnants of societies broken apart by an earlier pandemic disease (smallpox maybe). So everything we "know" about pre-contact native american societies that is not dug out of the ground (all the first hand reports from early contact -- cf "Conquest of New Spain" which is a great read and first hand journalism from a solider who was there with Cortez) may be wrong..
Anyway - just a minor addendum to your work and I'll look forward to reading more. By far the most informative post on slashdot I've ever been able to read..
Well said. If you haven't read Hunter's essay on a pre-presidential Carter speech, it's worth it. I can't find an online copy, but I believe it's in The Great Shark Hunt. I was able to find this nice video summary which is pretty well produced: http://www.youtube.com/watch?v=8SLeFZFTIco
Having spent the last two years as a political appointee in DC, I have to say that while it's not too late to go back, I am not seeing much in the political character (either of the nation or the politicians) that suggests the willingness to try. The circus feeds itself and as bubbles continue to collapse I think we'll see an increase in bickering and squabble not a decrease.. Sigh.
You don't really think that's the reason they don't have solar panels on the white house right? Carter put solar panels on the white house and Regan took them down. That's all you really need to know to understand the issues involved..
Good post. Errata -- I think you meant to write "EnterpriseDB's version of [[Postgres]] comes with PL/SQL compatibility, but only in a commercial product that lags behind the open-source releases--and buying from them just switches which vendor you're locked into."
Slashdot Mirror
Serious question: could you explain why SPDY is retarded? I'm actually curious what's wrong with, not knowing much about it other than having read some material when it first came out..
I love it when someone who actually knows something post on slashdot!
FTC has a do not call list that robocallers are obliged to wash their lists against. Not saying they all do it, but I put my numbers on that list and I get very few automated or human cold calls any more (and none from groups I haven't interacted with before).
That's very fair - good points. My thinking is that if AT&T can get me 3mbps DSL over copper for $30/month, the physical plant costs can't be the driver in pricing (and I'd think physical plant costs are probably lower for cable since the infrastructure is way less screwed up than copper pairs that the phone cos deal with)? Granted the phone co infrastructure is fully paid for so maybe that's an issue?
Of course DSL is less mbps but check out comcast pricing: you pay $60/month for 16mbps on cable. If you don't want 16mbps you can get 6mbps $50/month or 1.5mbps for $27/mo. The discontinuity in pricing could be related to loss leaders or physical plant (invariant) costs as you suggest, but given how cheap they are willing to bundle two or three services together, it smells like they are doing everything they can to encourage you to keep your cable subscription and not try to cut the cord and go internet only.
That makes sense if Comcast makes more money on cable subscribers than on internet subs? And why would they buy NBC if they weren't worried about cable subs cutting the cord? If they aren't in the content business then when their fat revenue from cable dries up, they become just another earthlink - big whoop.
I'm not saying I have the answers but the way they are running their business seems to suggest this line of thinking (and folks inside the FCC took this as common knowledge when I was there -- not that they necessarily have the inside track).. Curious what you think..
All votes today are not hand counted visually by a human (at least in California where I'm from), so I'm not sure what you're talking about, or rather I think this train has already left the station. Today, I physically mark a paper ballot, but the count is done by an optical scanner. Detection of fraud in that scanner can be done by random sampling, but I don't know for sure when or how this is actually done in CA.
I do know that random sampling as a practice can detect significant deviations in voter intention and recorded counts, and is used widely in ensuring fair voting in many countries around the world.
I agree with your sentiment that we shouldn't fool around with this system like we do with health records, banking or other so called "critical" services. This one critical service is far far more vulnerable and important than all the others..
The thing that chaps me is that I have to pay $60/month for internet from a cable company (and the phone company can't for whatever reason get optical or copper internet of any kind to me -- I live in downtown DC). For $70/month I could get basic cable and internet, but I don't want it. Since basic cable alone costs $40, it's clear that this pricing is b/c cable co's don't want internet service to be cheap, so they are artificially holding the price high, to retain cable subscribers, who are far, far more profitable than their internet customers (even when either customer is paying the same dollar value).
If I like to watch QVC I'll probably pay less!
Sorry - couldn't resist. There are some channels which pay to get access to viewers, whereas all the stations you want to watch are the other way around. If a station can make money by providing old movies which cost very little to license, then they are likely to make themselves free to cable, or even pay for a station. If content is expensive and/or if consumer demand is very high for something, then cable will pay to get it on their roster so as to have a competitive product.. Ugly business.
This is true for quantity as well. If I watch very little, I am subsidizing those who watch a lot with the all you can eat cable models. Netflix and Hulu are basically the same models - Netflix $8/month all the movies you can watch. Amazon movie rentals reward less viewing - you rent by title.. So this subsidy factor goes in a lot of directions.
Not sure if the analogy holds, but with internet service, Australia has had pay per byte models which some analysts argued (inside the FCC when I worked there) that this retarded AU's investments in internet infrastructure. US isn't that great for infrastructure either so I'm not sure I buy the argument, but the idea is that "all you can eat" is pro-innovation and development, and "pay as you go" offers more consumer choices, and lower average costs (at the cost of longer term stagnation in capital formation and investment in new services).
You're joking right? If not, the random sample is to check that all the votes that were counted by the machine are reflective of what the voters actually attempted to vote for (detecting mismatch between voter intention and what the machine is producing for counting). All votes are counted in any of these systems.
I don't fold my current voting sheet. I carry it from the booth to the reader wide out in the open where anyone can see it. If I want to stuff it in my shirt, or hold it against my chest I guess I could, but even then I have to mark both sides of the ballot so the under-the-shirt method is the only way I guess.
Anyway, this problem exists today at least in my county with regular paper ballots.
Yeah - really. The point as I understand UEFI is so that it can compete with desktop boxes and other baked in hardware where there is a relatively strong assurance that the stack of software that is currently running is "known good" from the operating system's perspective. This allows you to provide streaming services which are harder to rip and things like that. This in turn will probably make more movie and content companies make their stuff available on PC's.
This is just "trusted computing" if you remember that whole thing all over again. It's not about preventing you from running Linux, it's about preventing you from getting streaming movies on Linux, b/c it's harder to guarantee (to the movie's owner that) the stream is uncrackable..
Given that TC didn't work out so well (cracks on xbox360 and PS3) this may not play out the way they intend, but who knows maybe they'll finally get their dream of locking out people from these software stacks..
So companies who create/sell GPL aren't trying to benefit from freeriding on GPL? Like SugarCRM, Compiere or "old MySQL?" My experience is that (some) GPL software is released by companies who want to specifically prevent other companies from horning in on their business by re-using their codebase in a competitive way (no one but the original rights holder can release proprietary extensions or resell the product under other licenses). Not to start and GPL/academic license debate - but can you point to some companies that release academic license OSS who have business models like SugarCRM or the "old MySQL?" What am I missing?
Read this article to see how business-people see GPL: http://www.crmoutsiders.com/2010/07/15/some-thoughts-on-open-from-sugarcrm-ceo-larry-augustin/ -- GPL gives them a business model to keep others from selling the same thing that they sell to their customers (the core GPL version plus proprietary extensions).. Everyone else could only sell core GPL plus GPL extensions, and of course Sugar could reincorporate those gpl extensions into its core eliminating the competitor's differentiating features..
Wait - I recall reading a while back that Google isn't releasing 3.0-3.1 as open source but that they have said they *intend* to release future revs 3.3+ (or something like that) again as open? Am I misremembering? Is this a blip in the scheme of things or is Google in fact never planning to release another OSS version of Android? Anyone?
While I concede that Google's android app store is crappier than Apple's, it's possible that Google's goal with Android is very different from Apple's with the iPhone. Google may be trying to prevent someone like Apple or Microsoft from closing out the phone market in the way Microsoft did on the PC market years ago. If Google prevents that, then they will be able to continue to deliver search and ads on every platform.. I think this is probably a big reason why they are invested so heavily in Android: not because it is a profit center, but it is a strategic market force to keep their main lines of profit solid as the industry moves to tablets and phones from PCs and laptops..
In which case a somewhat crappier App environment while not desirable isn't that bad either for them.. So long as users are still adopting it and preventing a market unification around the iphone, which looked inevitable (at least to me) just a few years ago.
Love it - great post. I'm on this boat too. If your shit is so simple you just want a pile of data in a real fast container, use couch or mongo. If you want SQL b/c you need SQL use Pg. That is all.
Interesting. Possibly this is a case that demonstrates why BSD/MIT/Apache licenses are superior to GPL? When Oracle acquires a copyright title on a product such as MySQL released under GPL license, they can release closed source mods to the code and convert the community to a locked-in position, whereas no other player can do the same? Everyone but Oracle is "stuck" with a viral license now and so Oracle has a huge advantage in terms of controlling the community with proprietary extensions? Hmm.
I used Ruby/Rails with Postgres just fine on a very complex project, and I think that's a reasonably agile environment? The database migrations functions in Rails permitted management of state of database in dev, test, staging and production perfectly. It was actually the best setup I've ever had for that. You could see exactly what changes happened on each system, and you could very easily roll your code back to a particular date, roll the database back to the same date, and the structures and system data, lookups, etc would all get updated to reflect their state at that time. It made it so much easier to replicate a bug that might be happening in production, when your dev box was far in advance -- just check out the correct version, run your backward migration and you're testing on a system that's the same as production - no need to even create a new database or anything.
Postgres worked very well for all this. I think what you're saying is that *the way you like to manage these issues* doesn't work in Postgres, and that may very well be true. But that's not to say that you can't easily manage those issues in Postgres.
From what I read in the trades, the big reason that the EU allowed Oracle to keep MySQL during the acquisition was that Postgres exists.
The converse is also true - I know a few people who have iphone3 phones and who upgraded to iphone4 b/c the regular upgrades made their phones unusable. I suggested they jail break the iphone3 and go back to an earlier OS version (not even sure if this is possible not being an iphone user) but they weren't interested.
My Droid Eris is a lot slower than it used to be due to all the upgrades, but of course I do know that I can back-grade to an earlier OS version to get some speed back if it becomes unworkable for me.
But the carriers and manufacturers have plenty of incentive to over-upgrade your existing device to encourage you to go buy a new one. I wouldn't be surprised to start seeing a final "upgrade" release timed just about 2 years after the initial release so as to encourage users with their now too slow phones to go get new ones..
It's not burned into the silicon, it's loaded in the BIOS. Which implies it can be updated in the bios when vulns are found.
Great post - thanks. I'll review this stuff.. One thought on the 100x overpop comment -- you might be interested to read the book 1491 which discusses some reasonably mainstream archaeology theories that suggest north america may have supported populations approaching 100M prior to contact. And that very early contact spread disease through the Americas as a plague worse than the black death in Europe (33% population loss during that). So when Europeans first start actually walking around and looking for civilizations they see ragtag remnants of societies broken apart by an earlier pandemic disease (smallpox maybe). So everything we "know" about pre-contact native american societies that is not dug out of the ground (all the first hand reports from early contact -- cf "Conquest of New Spain" which is a great read and first hand journalism from a solider who was there with Cortez) may be wrong..
Anyway - just a minor addendum to your work and I'll look forward to reading more. By far the most informative post on slashdot I've ever been able to read..
Then watch this video: http://www.youtube.com/watch?v=HML5xhVCQwY
Well said. If you haven't read Hunter's essay on a pre-presidential Carter speech, it's worth it. I can't find an online copy, but I believe it's in The Great Shark Hunt. I was able to find this nice video summary which is pretty well produced: http://www.youtube.com/watch?v=8SLeFZFTIco
Having spent the last two years as a political appointee in DC, I have to say that while it's not too late to go back, I am not seeing much in the political character (either of the nation or the politicians) that suggests the willingness to try. The circus feeds itself and as bubbles continue to collapse I think we'll see an increase in bickering and squabble not a decrease.. Sigh.
You don't really think that's the reason they don't have solar panels on the white house right? Carter put solar panels on the white house and Regan took them down. That's all you really need to know to understand the issues involved..
Good post. Errata -- I think you meant to write "EnterpriseDB's version of [[Postgres]] comes with PL/SQL compatibility, but only in a commercial product that lags behind the open-source releases--and buying from them just switches which vendor you're locked into."