They sold me "20 MB/s" cable service. That suggests to me (and the rest of the plaintiffs) that it should be 20 MegaBytes per second. VM claim (of course) that it's 20 MegaBits per second. Looking at thier site the b is clearly lowercase. Afaict network data rate has always been measured in megabits per second represented by a lower case b. Every other ISP i'm aware of measures in the same way. Do you have any evidence that they used a capital B on the material that was used to market the service to you?
They then apply "STM" - Subscriber Traffic Management. The effect of this is that if you download anything for just 20 minutes in any day, your data rate is reduced to 25% of your rated speed... Now that is IMO unreasonable.
Freecell games are built from complete decks by suit into the stacks that you start with. What that means is that every game is solvable because it started in a solved state and was deconstructed into the puzzle. bullshit
the cards are arranged by the output of a prng. The game number is the seed of that prng. This certainly has the potential to produce an unsolvable layout though in practice it very rarely does (iirc of the 32000 games availible in the original MS version of freecell only one is belived to be unsolvable).
Yeah, because Debian's 5(?) year gap between stable versions Debian did drop the ball with sarge getting stuck in thier longest release cycle ever at just the time that desktop linux was getting a lot more friendly (in particular the rise of X autoconfiguration) but even that cycle was only 3 years.
Every other debian release cycle has been under 2 years.
MS was already a near monopoly by the time XP came along. I don't think windows popularity was dented much by XPs long lifetime though the fact that vista when it finally arrived was underwhelming to say the least has made a lot of people unhappy.
Release cycles are a compromise, too fast and you either end up with users pissed off at having to upgrade all the time or an insane suport load. Too slow and at least in the linux world you get people being pissed off at not being able to run the latest software (this is not an issue for MS because there are not multiple competing windows distributions, targetting windows means targetting official releases of windows).
I'm not convinced trying to force the whole free software community to try and sync up to a common release cycle is a good idea. Sure it may mean that end users get new features a bit sooner than they would otherwise and it will make it easier to run an enterprise distribution on a low budget (which is why ubuntu wants it) but it will also force many projects into release cycles that do not fit with them.
I don't think microsoft tetris for windows has that behaviour though.
In any case the game still controls the schedule. If you don't keep spinning that peice (or pause but that tends to require moving your fingers away from the main controls limiting when you can safely do it) it will drop.
There's too much luck involved and not enough skill. In some there is a lot of luck but in freecell all the cards positions are known at the start of the game and almost every game is solvable.
Nintendo was dominating the handheld video game market long before the DS.
OTOH the touch screen in the DS did open up new categories of games which appealed to new categories of users. I suspect we will never know if that was intentional or just a good side effect for nintendo.
Most early touchscreen systems I encountered (mostly "interactive" exibits in museums) took a very long time to give feedback to a touch. With no physical feedback fast visual feedback becomes even more vital to making the system feel responsive.
As you say you would need much processor power to get a quick scan of the screen. I suspect the problems were either on the coordinate-object translation or on the display side.
A finger is a rather clumsy interface device compared to the pinpoint precision offered by a mouse. A stylus helps significantly there though and while it is not quite as good as a mouse I would imagine in most cases it would give sufficiant precision to drive a conventional windows like GUI easilly.
Obviously something is blocking the loading/rendering in all open tabs when this is happening. Having seen the effect I don't think that is what is going on because while this is happening I can load pages from other sites in other tabs with no problems.
IIRC firefox has by default a relatively low limit on the number of connections to a single server (iirc this can be changed through about:config). I suspect that cap is being hit by the "stuck" connections preventing further downloads from that server.
Afaict ubuntu is trying to do a LTS release on a relatively (compared to redhat/novell) small budget by avoiding having to backport security fixes themselves. I understand why they are doing this (they want to break into the enterprise market becaue that is where the real money is) but I wonder if it will backfire on them.
pidgins standard msn protocol plugin is in very poor shape at the moment and thier experimental MSNP14 code is almost as bad. Lukilly there is a third party one (msn-pecan) which seems a lot better.
Also in general moving to a multiprotocol client means taking a large feature hit. Whether this matters to you will vary depending on whether you use those advanced features or only care about basic messaging.
I have worked for several successful software companies and this is how it works at all of them. The customer wants stable code on schedule and the release process is designed with that understanding. Right, you structure your release process arround what your customers wan't, a perfectly reasonable thing to do. You probablly also try to pressure your suppliers to work to your schedule with varying degrees of success depending on how the ammount you pay them compares to their overall size.
The fact is that canonical/ubuntu is a relatively small operation, essentially it is the plaything of one moderately rich guy. But shuttleworth while being rich is nowhere near rich enough to fund a redhat size operation out of his own pocket.
They want to move into the enterprise market (because that is where the money lies)but they don't really have the rescources to do so. So essentially they are trying to make running an enterprise linux distro either by forming a distro cartel which will put pressure on the upstream projects to work on thier terms.
redhat and novell would be mad to accept such a proposal. They control the linux enterprise market and it is obiously in thier interests that it stays hard for upstarts like canonical to get a footing in that market.
redhat gets a lot of income from being the dominant enterprise linux vendor. They use this money to pay people to develop on major open source projects. Doing so brings them a number of benifits.
* They have people intimately familiar with the code on thier payroll. ** If they need a patch to an older version they can get it written immediately, not have to wait ages while an unfamiliar coder learns the codebases and works out how to backport the fix. ** If someone calls them for support or custom changes they can put them onto someone intimately familiar with the codebase. * They have influence with upstream over release/support schedules.
Shuttleworth is pretty rich (the wikipedia article says he sold thwate for the equivilent of just over half a billion USD) but canonical is way way smaller than redhat.
Ubuntu has got a load of ordinary desktop users. but such users aren't very profitable. Thier long term future depends on being able to break into the more lucrative "enterprise" market. But enterprise customers depend long support cycles and ubuntu doesn't have the rescourses to provide that without upstreams help.
Which will put even more (ubuntus runaway popularity is already exerting a lot) pressure on app developers to sync to distro release schedules even if it is destructive to thier other goals.
Unfortunately in many countries the cellphone networks have set up thier price plans such that (unless you ave a very light user who is better off on a pay as you go plan) you pay them for a new cellphone every year or so whether you actually take said phone or not,
While pay as you go users don't get "free" phones from the network there is often still some subsidy of the phone price and of course a lot of pay as you go users are using phones handed down by contract users who have since received a new "free" phone.
The result is of course that phones which act in ways that are against the networks interests do not become common.
Yes it has been tried, the trouble is the predators don't always eat what you want them to eat, sometimes they preffer the local wildlife to the species you are trying to control.
2K and Me were the first versions of windows with decent support for USB mass storage. USB on 95 existed in the last OEM service releases but there was very little driver support. 98 seemed to require (there may have been third party drivers that added more general support but if there were I never found them) a specific driver for each different flash stick you wanted to use (which often aren't availible for new flash sticks anymore).
I know at uni there are lots of old PCs controlling various bits of equipment. Often theese machines are running very old operating systems so USB mass storage is not an option (MS really only got USB mass storage right with ME/2K , with 98 you could use them but only with a manufacturer specific driver or some third party kludge, with 95 or NT4 you don't have much chance).
Well it looks like only about 200 megs is actually being used by applications. The buffers categorty is memory that linux is using to buffer data from disk and can free up if an application needs it.
The point is we normally measure cars speeds in a measurement that is good for relating journey distances to journey times. Not measurements that are good for relating thinking times to distance travelled before reacting.
It isn't as bad as some people are making out. If you have IE7 installed when you install SP3 you can't go back to IE6 afterwards but SP3 does not force you to upgrade to IE7.
And as you imply if you have to run an old insecure browser for some specific task using a VM is probablly a good idea.
XP without service packs does have a built in firewall that is adequate for the purpose of protecting a box while installing updates. You just have to remember to turn it on.
SP2 was far from fine, it broke a lot of popular software including the version of nero I had at the time and attempting to install it on an unclean machine would often completely break the windows install.
Slashdot Mirror
If you have a good network connection to a *nix box you could always run an X server on your windows box and then run gnome-terminal using remote X.
Someone may have just forgotten to block an internal routing advertisement from being propogated to the outside or something.
They sold me "20 MB/s" cable service. That suggests to me (and the rest of the plaintiffs) that it should be 20 MegaBytes per second. VM claim (of course) that it's 20 MegaBits per second.
Looking at thier site the b is clearly lowercase. Afaict network data rate has always been measured in megabits per second represented by a lower case b. Every other ISP i'm aware of measures in the same way. Do you have any evidence that they used a capital B on the material that was used to market the service to you?
They then apply "STM" - Subscriber Traffic Management. The effect of this is that if you download anything for just 20 minutes in any day, your data rate is reduced to 25% of your rated speed...
Now that is IMO unreasonable.
Freecell games are built from complete decks by suit into the stacks that you start with. What that means is that every game is solvable because it started in a solved state and was deconstructed into the puzzle.
bullshit
the cards are arranged by the output of a prng. The game number is the seed of that prng. This certainly has the potential to produce an unsolvable layout though in practice it very rarely does (iirc of the 32000 games availible in the original MS version of freecell only one is belived to be unsolvable).
Yeah, because Debian's 5(?) year gap between stable versions
Debian did drop the ball with sarge getting stuck in thier longest release cycle ever at just the time that desktop linux was getting a lot more friendly (in particular the rise of X autoconfiguration) but even that cycle was only 3 years.
Every other debian release cycle has been under 2 years.
MS was already a near monopoly by the time XP came along. I don't think windows popularity was dented much by XPs long lifetime though the fact that vista when it finally arrived was underwhelming to say the least has made a lot of people unhappy.
Release cycles are a compromise, too fast and you either end up with users pissed off at having to upgrade all the time or an insane suport load. Too slow and at least in the linux world you get people being pissed off at not being able to run the latest software (this is not an issue for MS because there are not multiple competing windows distributions, targetting windows means targetting official releases of windows).
I'm not convinced trying to force the whole free software community to try and sync up to a common release cycle is a good idea. Sure it may mean that end users get new features a bit sooner than they would otherwise and it will make it easier to run an enterprise distribution on a low budget (which is why ubuntu wants it) but it will also force many projects into release cycles that do not fit with them.
I don't think microsoft tetris for windows has that behaviour though.
In any case the game still controls the schedule. If you don't keep spinning that peice (or pause but that tends to require moving your fingers away from the main controls limiting when you can safely do it) it will drop.
There's too much luck involved and not enough skill.
In some there is a lot of luck but in freecell all the cards positions are known at the start of the game and almost every game is solvable.
Nintendo was dominating the handheld video game market long before the DS.
OTOH the touch screen in the DS did open up new categories of games which appealed to new categories of users. I suspect we will never know if that was intentional or just a good side effect for nintendo.
Most early touchscreen systems I encountered (mostly "interactive" exibits in museums) took a very long time to give feedback to a touch. With no physical feedback fast visual feedback becomes even more vital to making the system feel responsive.
As you say you would need much processor power to get a quick scan of the screen. I suspect the problems were either on the coordinate-object translation or on the display side.
A finger is a rather clumsy interface device compared to the pinpoint precision offered by a mouse.
A stylus helps significantly there though and while it is not quite as good as a mouse I would imagine in most cases it would give sufficiant precision to drive a conventional windows like GUI easilly.
Obviously something is blocking the loading/rendering in all open tabs when this is happening.
Having seen the effect I don't think that is what is going on because while this is happening I can load pages from other sites in other tabs with no problems.
IIRC firefox has by default a relatively low limit on the number of connections to a single server (iirc this can be changed through about:config). I suspect that cap is being hit by the "stuck" connections preventing further downloads from that server.
Afaict ubuntu is trying to do a LTS release on a relatively (compared to redhat/novell) small budget by avoiding having to backport security fixes themselves. I understand why they are doing this (they want to break into the enterprise market becaue that is where the real money is) but I wonder if it will backfire on them.
pidgins standard msn protocol plugin is in very poor shape at the moment and thier experimental MSNP14 code is almost as bad. Lukilly there is a third party one (msn-pecan) which seems a lot better.
Also in general moving to a multiprotocol client means taking a large feature hit. Whether this matters to you will vary depending on whether you use those advanced features or only care about basic messaging.
I have worked for several successful software companies and this is how it works at all of them. The customer wants stable code on schedule and the release process is designed with that understanding.
Right, you structure your release process arround what your customers wan't, a perfectly reasonable thing to do. You probablly also try to pressure your suppliers to work to your schedule with varying degrees of success depending on how the ammount you pay them compares to their overall size.
The fact is that canonical/ubuntu is a relatively small operation, essentially it is the plaything of one moderately rich guy. But shuttleworth while being rich is nowhere near rich enough to fund a redhat size operation out of his own pocket.
They want to move into the enterprise market (because that is where the money lies)but they don't really have the rescources to do so. So essentially they are trying to make running an enterprise linux distro either by forming a distro cartel which will put pressure on the upstream projects to work on thier terms.
redhat and novell would be mad to accept such a proposal. They control the linux enterprise market and it is obiously in thier interests that it stays hard for upstarts like canonical to get a footing in that market.
Agreed,
redhat gets a lot of income from being the dominant enterprise linux vendor. They use this money to pay people to develop on major open source projects. Doing so brings them a number of benifits.
* They have people intimately familiar with the code on thier payroll.
** If they need a patch to an older version they can get it written immediately, not have to wait ages while an unfamiliar coder learns the codebases and works out how to backport the fix.
** If someone calls them for support or custom changes they can put them onto someone intimately familiar with the codebase.
* They have influence with upstream over release/support schedules.
Shuttleworth is pretty rich (the wikipedia article says he sold thwate for the equivilent of just over half a billion USD) but canonical is way way smaller than redhat.
Ubuntu has got a load of ordinary desktop users. but such users aren't very profitable. Thier long term future depends on being able to break into the more lucrative "enterprise" market. But enterprise customers depend long support cycles and ubuntu doesn't have the rescourses to provide that without upstreams help.
Which will put even more (ubuntus runaway popularity is already exerting a lot) pressure on app developers to sync to distro release schedules even if it is destructive to thier other goals.
Unfortunately in many countries the cellphone networks have set up thier price plans such that (unless you ave a very light user who is better off on a pay as you go plan) you pay them for a new cellphone every year or so whether you actually take said phone or not,
While pay as you go users don't get "free" phones from the network there is often still some subsidy of the phone price and of course a lot of pay as you go users are using phones handed down by contract users who have since received a new "free" phone.
The result is of course that phones which act in ways that are against the networks interests do not become common.
Yes it has been tried, the trouble is the predators don't always eat what you want them to eat, sometimes they preffer the local wildlife to the species you are trying to control.
2K and Me were the first versions of windows with decent support for USB mass storage. USB on 95 existed in the last OEM service releases but there was very little driver support. 98 seemed to require (there may have been third party drivers that added more general support but if there were I never found them) a specific driver for each different flash stick you wanted to use (which often aren't availible for new flash sticks anymore).
I know at uni there are lots of old PCs controlling various bits of equipment. Often theese machines are running very old operating systems so USB mass storage is not an option (MS really only got USB mass storage right with ME/2K , with 98 you could use them but only with a manufacturer specific driver or some third party kludge, with 95 or NT4 you don't have much chance).
Well it looks like only about 200 megs is actually being used by applications. The buffers categorty is memory that linux is using to buffer data from disk and can free up if an application needs it.
I still think it is rather bloated though.
The point is we normally measure cars speeds in a measurement that is good for relating journey distances to journey times. Not measurements that are good for relating thinking times to distance travelled before reacting.
It isn't as bad as some people are making out. If you have IE7 installed when you install SP3 you can't go back to IE6 afterwards but SP3 does not force you to upgrade to IE7.
And as you imply if you have to run an old insecure browser for some specific task using a VM is probablly a good idea.
XP without service packs does have a built in firewall that is adequate for the purpose of protecting a box while installing updates. You just have to remember to turn it on.
SP2 was far from fine, it broke a lot of popular software including the version of nero I had at the time and attempting to install it on an unclean machine would often completely break the windows install.