But what is the definition of "fail"? And what conditions are those drives subject to? are they run continuousy? are they subjected to power removal without warning?
How is that calculated? in particular do drives that corrupt data silently or worse lock the user out of their data requiring a hard reset to make the drive work again count or do only drives that actually fail permanently and get RMA'd count?
Whether the summary is "wrong" depends on whether you read "longest any human has been off Earth" as single visit time or total time across multiple visits.
I agree with your general principle but I have to pick some nits in your case of ethernet.
which seems to have an unbroken chain of compatibility, in fact they only maintain compatibility a single generation back, and discard what came further before.
Well depends a bit on how you count "generations" but if you count each physical layer that was in common use for end user machines as a "generation" then.
10BASE-5 original commercial ethernet. Ususally not connected directly to end-systems but connected by transcievers. 10BASE-2 cheapened version usually used with built-in transcivers. I don't think interconnecting 10BASE-2 and 10BASE-5 directly was officially reccomended but from what I can gather it did work in practice. 10BASE-T totally new physical layer but many network cards and hubs hubs had coax or AUI ports as well. 100BASE-TX new electrical specification on existing connector. Most network cards and switches nearly always supported 10BASE-T as well but there were some single speed hubs out there (a pure hub is single speed by definition, a device that supports multiple speeds at the same time must support some form of bridiging). 1000BASE-T another new electrical specification on an existing connector. Hubs are nonexistent in the market place and network cards and switches nearly always support 10BASE-T and 100BASE-T. 10GBASE-T hasn't become a widespread end-user connection yet so it's not clear how much backwards comaptibility consumer cards will implement (AIUI current server cards implement compatibility with 1000BASE-T and 100BASE-TX but not 10BASE-T).
So I make that between 0 and 2 generations of direct compatibility depending on the particular device. However most of the compatibility breaks were a LONG time ago.
Notice that the GigE cards don't come with AUI or BNC connectors?
But they do come with support for both 10BASE-T and 100BASE-T.
Notice that there are no GigE hubs?
Pure hubs* were single speed while switches allow interconnection of different speeds. So the abandoning of hubs in favour of switches was a good thing from a backwards compatibility perspective.
That GigE doesn't work over the CAT-3 cabling that was installed en-mass way back when?
True and it's annoying that ethernet autonegotiation doesn't take any account of the condition of the cable, only the capabilities of the devices.
The great thing with ethernet is that while the physical layer has changed because the higher levels have remained much the same for a long time one can still use old equipment reasonablly easilly. If I grab an old peice of kit with an AUI port on it, plug a 10base-T transciever into it and connect a cable between that transciver and my gigabit switch then it doesn't know that the transciver I plugged in is a 10BASE-T one not a 10BASE-5 one. Nor does it know that the devices on the other side of the switch are operating at gigabit speeds.
* There were some multi-speed devices advertised as hubs that interconnected 10base-T and 100base-T while being cheaper than a proper switch (for example they could act as a 10 megabit hub, a 100 megabit hub and a 2-port birdge).
The DS series has more in common with old-school consoles than with phones/tablets/PCs/modern consoles. Games largely hit the hardware directly rather than going through abstraction layers.
Remember the DS series is 10 years old so the design descisions were made in the context of what hardware was like 10+ years ago. Furthermore the DS was compatible with GBA games so some of it's architecture was influenced by design decisions made even earlier.
They can add new features for new games but it would be very difficult to change to a modern design without a compatibility break.
Afaict this means that older DS games are basically unusuable on modern wifi systems. In addition to the versions of wifi itself there is the issue of lack of support for modern wireless security protocols.
You will probablly be able to get an external box that will allow the printer to be used on the new wifi but I expect it will cost a lot more than $2 and you may lose some functionality (especially if the printer doesn't have ethernet).
Where things get messy is you can legitimately turn real money into game money by buying plex with real money and selling them for ingame money and you can turn in-game money into extensions of your subscription by buying plex ingame and then using them.
But there is no approved way to turn ingame money back into real money. They can't totally prevent people doing it but it's a violation of the terms of service and they can and will take action if they catch people doing it.
Is it realistic to expect Windows/Mac OS X for ARM in their desktop versions in the near future?
What I find really curious is that MS did all the work needed to put a full version of windows on arm.
Then they turned it into a crippled peice of shit with artificial restricitions. No third party desktop apps, no ability to join corporate domains, third party developers pushed hard into using the windows store with it's apple-like fees (AIUI there are ways to load your own metro apps without using the store but they aren't exactly user friendly).
Depeneds what exactly the "storage box" is doing with the data.
If it is doing block level deduplication then ram starts to become very important since you really want to keep the deduplication tables in ram. The freenas guys reccomend 5GB of ram per terabyte of storage for ZFS deduplication.
If it's serving up the same files repeately then more ram means more chance that those files will be cached in memory rather than having to be read from relatively slow storage devices.
I was under the impression that in at least some cases the same turbines could be used with either natural gas or kerosense depending on which was cheaper at the time (natural gas is usually a lot cheaper than oil based fuel oils but because it's harder to store and transport it can fluctuate a lot).
Fundamentally they have to buy the key components from Intel (processor and PCH) and ATI or NVIDIA (GPU) and connect them together in pretty much the standard way. They may have a bit more leverage to request design tweaks than some vendors and they do use custom board layouts to fit their form factor demands but I highly doubt they have enough leverage to specify a major redesign of the core comonents.
I suspect this move will win them some favour among debian developers, maybe it will get a few patches accepted quicker or make some debian developers more likely to consider taking a job working for them. I find it highly unlikely that it would win them enough favour to have any signficant impact on the project as a whole. The hard line "I only use FOSS" types aren't going to be swayed by a few "free as in beer" games.
You missed DOTA 2, Day of defeat, team fortress classic and deathmatch classic (granted you might consider some of those part of the "half life/counterstrike series").
When I look on steam the "valve complete pack" currently costs £50. Of course if you wait for a sale you can probablly pick it up much cheaper than that.
which begs the question: Who is considered a developer?
Debian Developer is a well defined status with full upload rights and voting rights. The application process is also documented but well basically it consists of
1: get your key signed by a couple of existing debian developers (in a pinch one debian developer and one 2: find an existing dd to advocate for you 3: make your application 4: wait to be assigned an application manager 5: go through a questioning/testing process with the application manager 6: wait for final approval, account creations etc from front desk, DAM and keyring maint
Nowadays it's generally reccomended to go for the lower status of "Debian Maintainer" (restricted upload rights, no voting rights) first and then move on to applying for Debian Developer (I went straight to DD myself so it IS still possible to go directly in the right circumstances but it's not considered the normal route anymore).
The process of becoming a Debian Developer can take quite some time both in terms of overall process length and the ammount you will have to learn about debian and the contributions you will have to demonstrate you have made to pass it. If you are serious about contributing to Debian then it's worth it, if you are just doing it for the handful of goodies (the ones i'm aware of are lwn subscription, ghandi.net hosting discount, and now valve games) that are given out debian developers then you are wasting your time.
I think the "or at least the core product benefit" is the key part. We replaced horses and carriges with internal combustion engines but the "core product benefit" (hauling stuff arround without using our own muscle power) was still there.
Similarly facebook may get replaced with something else in time but I suspect there will always be services that provide similar "friend management and rediscovery" services, messaging services, login services and so-on. It will be an uphill struggle because of the network effects but it can happen, particulally when you talk about timescales of generations.
There are third party archives (both archive.org and dedicated archiving products) of at least some of the content but geocities itself has closed down.
Slashdot Mirror
But what is the definition of "fail"?
And what conditions are those drives subject to? are they run continuousy? are they subjected to power removal without warning?
reliability rate
How is that calculated? in particular do drives that corrupt data silently or worse lock the user out of their data requiring a hard reset to make the drive work again count or do only drives that actually fail permanently and get RMA'd count?
Whether the summary is "wrong" depends on whether you read "longest any human has been off Earth" as single visit time or total time across multiple visits.
If you look at something like Ethernet,
I agree with your general principle but I have to pick some nits in your case of ethernet.
which seems to have an unbroken chain of compatibility, in fact they only maintain compatibility a single generation back, and discard what came further before.
Well depends a bit on how you count "generations" but if you count each physical layer that was in common use for end user machines as a "generation" then.
10BASE-5 original commercial ethernet. Ususally not connected directly to end-systems but connected by transcievers.
10BASE-2 cheapened version usually used with built-in transcivers. I don't think interconnecting 10BASE-2 and 10BASE-5 directly was officially reccomended but from what I can gather it did work in practice.
10BASE-T totally new physical layer but many network cards and hubs hubs had coax or AUI ports as well.
100BASE-TX new electrical specification on existing connector. Most network cards and switches nearly always supported 10BASE-T as well but there were some single speed hubs out there (a pure hub is single speed by definition, a device that supports multiple speeds at the same time must support some form of bridiging).
1000BASE-T another new electrical specification on an existing connector. Hubs are nonexistent in the market place and network cards and switches nearly always support 10BASE-T and 100BASE-T.
10GBASE-T hasn't become a widespread end-user connection yet so it's not clear how much backwards comaptibility consumer cards will implement (AIUI current server cards implement compatibility with 1000BASE-T and 100BASE-TX but not 10BASE-T).
So I make that between 0 and 2 generations of direct compatibility depending on the particular device. However most of the compatibility breaks were a LONG time ago.
Notice that the GigE cards don't come with AUI or BNC connectors?
But they do come with support for both 10BASE-T and 100BASE-T.
Notice that there are no GigE hubs?
Pure hubs* were single speed while switches allow interconnection of different speeds. So the abandoning of hubs in favour of switches was a good thing from a backwards compatibility perspective.
That GigE doesn't work over the CAT-3 cabling that was installed en-mass way back when?
True and it's annoying that ethernet autonegotiation doesn't take any account of the condition of the cable, only the capabilities of the devices.
The great thing with ethernet is that while the physical layer has changed because the higher levels have remained much the same for a long time one can still use old equipment reasonablly easilly. If I grab an old peice of kit with an AUI port on it, plug a 10base-T transciever into it and connect a cable between that transciver and my gigabit switch then it doesn't know that the transciver I plugged in is a 10BASE-T one not a 10BASE-5 one. Nor does it know that the devices on the other side of the switch are operating at gigabit speeds.
* There were some multi-speed devices advertised as hubs that interconnected 10base-T and 100base-T while being cheaper than a proper switch (for example they could act as a 10 megabit hub, a 100 megabit hub and a 2-port birdge).
The DS series has more in common with old-school consoles than with phones/tablets/PCs/modern consoles. Games largely hit the hardware directly rather than going through abstraction layers.
Remember the DS series is 10 years old so the design descisions were made in the context of what hardware was like 10+ years ago. Furthermore the DS was compatible with GBA games so some of it's architecture was influenced by design decisions made even earlier.
They can add new features for new games but it would be very difficult to change to a modern design without a compatibility break.
Afaict this means that older DS games are basically unusuable on modern wifi systems. In addition to the versions of wifi itself there is the issue of lack of support for modern wireless security protocols.
You will probablly be able to get an external box that will allow the printer to be used on the new wifi but I expect it will cost a lot more than $2 and you may lose some functionality (especially if the printer doesn't have ethernet).
In other words, if that wheel is too small and you spin it too fast (to get to that 1g you want), you'd be nauseated to the extreme.
Would you stay nauseated though or would you get used to it?
Where things get messy is you can legitimately turn real money into game money by buying plex with real money and selling them for ingame money and you can turn in-game money into extensions of your subscription by buying plex ingame and then using them.
But there is no approved way to turn ingame money back into real money. They can't totally prevent people doing it but it's a violation of the terms of service and they can and will take action if they catch people doing it.
Is it realistic to expect Windows/Mac OS X for ARM in their desktop versions in the near future?
What I find really curious is that MS did all the work needed to put a full version of windows on arm.
Then they turned it into a crippled peice of shit with artificial restricitions. No third party desktop apps, no ability to join corporate domains, third party developers pushed hard into using the windows store with it's apple-like fees (AIUI there are ways to load your own metro apps without using the store but they aren't exactly user friendly).
Depeneds what exactly the "storage box" is doing with the data.
If it is doing block level deduplication then ram starts to become very important since you really want to keep the deduplication tables in ram. The freenas guys reccomend 5GB of ram per terabyte of storage for ZFS deduplication.
If it's serving up the same files repeately then more ram means more chance that those files will be cached in memory rather than having to be read from relatively slow storage devices.
I believe you typically also have to have been a DM (Debian Maintainer) for at least 6 months prior.
It's possible to become a DD without being a DM first. Of course you will have to demonstrate significant contributions to Debian in some other way.
edit: according to the followup on debian-devel DMs are in despite the original announcement only mentioning DDs.
Is it normal in china to write news about a rover as if the rover was writing it?
I was under the impression that in at least some cases the same turbines could be used with either natural gas or kerosense depending on which was cheaper at the time (natural gas is usually a lot cheaper than oil based fuel oils but because it's harder to store and transport it can fluctuate a lot).
They just split the company yet again
According to the article you linked they haven't actually done it yet, just announced that they intend to do it.
Fundamentally they have to buy the key components from Intel (processor and PCH) and ATI or NVIDIA (GPU) and connect them together in pretty much the standard way. They may have a bit more leverage to request design tweaks than some vendors and they do use custom board layouts to fit their form factor demands but I highly doubt they have enough leverage to specify a major redesign of the core comonents.
But, Apple started making PCs in 1976, 4 years before HP according to HPs own timeline referenced in the summary.
According to their timeline HP were advertising a device as a "personal computer" in 1968
I suspect this move will win them some favour among debian developers, maybe it will get a few patches accepted quicker or make some debian developers more likely to consider taking a job working for them. I find it highly unlikely that it would win them enough favour to have any signficant impact on the project as a whole. The hard line "I only use FOSS" types aren't going to be swayed by a few "free as in beer" games.
It seems they have banned discriminating against old people but not discriminating against young people.
You missed DOTA 2, Day of defeat, team fortress classic and deathmatch classic (granted you might consider some of those part of the "half life/counterstrike series").
When I look on steam the "valve complete pack" currently costs £50. Of course if you wait for a sale you can probablly pick it up much cheaper than that.
It is also open to debian maintainers
Hmm, the announcement on Debian-devel-announce only mentioned Debian Developers not Debian Maintainers.
It appears that first you must get on an airplane
Whether an airplane is needed rather depends on where you live. In your case it looks like a car would suffice.
https://wiki.debian.org/Keysig...
which begs the question: Who is considered a developer?
Debian Developer is a well defined status with full upload rights and voting rights. The application process is also documented but well basically it consists of
1: get your key signed by a couple of existing debian developers (in a pinch one debian developer and one
2: find an existing dd to advocate for you
3: make your application
4: wait to be assigned an application manager
5: go through a questioning/testing process with the application manager
6: wait for final approval, account creations etc from front desk, DAM and keyring maint
Nowadays it's generally reccomended to go for the lower status of "Debian Maintainer" (restricted upload rights, no voting rights) first and then move on to applying for Debian Developer (I went straight to DD myself so it IS still possible to go directly in the right circumstances but it's not considered the normal route anymore).
The process of becoming a Debian Developer can take quite some time both in terms of overall process length and the ammount you will have to learn about debian and the contributions you will have to demonstrate you have made to pass it. If you are serious about contributing to Debian then it's worth it, if you are just doing it for the handful of goodies (the ones i'm aware of are lwn subscription, ghandi.net hosting discount, and now valve games) that are given out debian developers then you are wasting your time.
I think the "or at least the core product benefit" is the key part. We replaced horses and carriges with internal combustion engines but the "core product benefit" (hauling stuff arround without using our own muscle power) was still there.
Similarly facebook may get replaced with something else in time but I suspect there will always be services that provide similar "friend management and rediscovery" services, messaging services, login services and so-on. It will be an uphill struggle because of the network effects but it can happen, particulally when you talk about timescales of generations.
is geocities still around?
There are third party archives (both archive.org and dedicated archiving products) of at least some of the content but geocities itself has closed down.