Date disaster in 30 year's time?
in [UK Mac]
|
Prev: iTunes - sync Gmail contacts ?
Next: Bookmark Syncing
From: Chris Ridd on 4 Mar 2010 01:32 On 2010-03-04 00:37:51 +0000, Richard Tobin said: > In article <1815951159289344162.828723chrisridd-mac.com(a)news.individual.net>, > Chris Ridd <chrisridd(a)mac.com> wrote: > >>> Unix uses the number of seconds since Jan 1 1970. It traditionally >>> stores this in a 32-bit signed integer, which will overflow in >>> 2038. Presumably this will have changed to a 64-bit integer long >>> before then. > >> It already has - see those 64-bit apps? > > I checked before posting, but unfortunately I checked on a Leopard > system. It is indeed 64 bits when compiled in default mode on Snow > Leopard. > >> There's no way to properly allow >> 32-bit apps to use bigger time_ts while keeping binary compatibility, so >> 64 is the way to go. > > It's a pity Apple didn't fix a few things like this when they switched > to x86, when there was no binary compatibility to maintain. Yes, they could have done that. But perhaps they felt that everyone would be 64-bit by then so there was no point? Also it would break compatibility with 32-bit PPC in perhaps too-subtle ways. Casting time_t to int and back would give different results for instance. Bad programming, but probably not uncommon... > If there are still x86 32-bit systems around by then (which wouldn't > surprise me - the 80386 was introduced 25 years ago, and there's only > 28 years left), a workaround would be to change the interpretation of > large negative times, making the range (say) 1950-2087, instead of > 1901-2038. The main problem is that time_t values of -1 are defined to indicate an error (which is the only reason time_t is signed), so you'd need to figure out a way to avoid that. -- Chris
From: Peter Ceresole on 4 Mar 2010 02:36 Chris Ridd <chrisridd(a)mac.com> wrote: > Yes, they could have done that. But perhaps they felt that everyone > would be 64-bit by then so there was no point? Don't you think they're fundamentally right? 28 years from now... There will be a (very) few people running 32-bit Mac systems, for historical reasons, or because they are determined to be able to read files generated by old applications, but the date function shouldn't be a major problem. It will have wrapped round to 1900-something but that's unlikely to prove a fundamental problem for that kind of use. Anybody generating files *now* that need to function operationally then will become aware of the problem quite soon- if they aren't already- and will set about copying and updating their important files to a 64-bit (or more) system. If they don't, they deserve to fail. And they have quite a bit of time to do it. -- Peter
From: D.M. Procida on 4 Mar 2010 03:50 Peter Ceresole <peter(a)cara.demon.co.uk> wrote: > Anybody generating files *now* that need to function operationally then > will become aware of the problem quite soon- if they aren't already- and > will set about copying and updating their important files to a 64-bit > (or more) system. If they don't, they deserve to fail. This is the kind of thing that makes me doubt that digital archives of photographs will last longer than physical ones. Daniele
From: J. J. Lodder on 4 Mar 2010 04:25 D.M. Procida <real-not-anti-spam-address(a)apple-juice.co.uk> wrote: > Peter Ceresole <peter(a)cara.demon.co.uk> wrote: > > > Anybody generating files *now* that need to function operationally then > > will become aware of the problem quite soon- if they aren't already- and > > will set about copying and updating their important files to a 64-bit > > (or more) system. If they don't, they deserve to fail. > > This is the kind of thing that makes me doubt that digital archives of > photographs will last longer than physical ones. Why? A .jpeg will sill be a .jpeg hundred years hence, and there is no reason at all why it wouldn't be possible to display it, using a suitable app, if the medium is still readable, Jan PS If long term survival bothers you it might be a good idea to include 10% par2 with your images.
From: Peter Ceresole on 4 Mar 2010 04:26
D.M. Procida <real-not-anti-spam-address(a)apple-juice.co.uk> wrote: > This is the kind of thing that makes me doubt that digital archives of > photographs will last longer than physical ones. I think they will last in the hands of their current owners- for instance, I know that I will keep my own pictures and transfer them from machine to machine as I progress, until I go phut. And reading the image formats should be okay; I can see JPEGs being decipherable pretty well indefinitely. Software will survive much longer than machinery, because the machinery wears out and is expensive to replace, whereas that kind of software doesn't and isn't. But it does require archives to transfer and safeguard their digital data as carefully as they protect and maintain their physical stores. Clearly, the good ones will do so. But real life experience says that many will not. And physical existence doesn't mean that material survives; I've given the example of skip-loads of what would be sociologically fascinating material being dumped because we changed buildings (Lime Grove to TV Centre) and there was nowhere to keep the stuff. And no money to build anew. Some of the more 'important' material has been saved, and is brought out in retrospective programmes- I can still recognise some of the cuts I made as a film editor cropping up in films about the '60s. But this weeding process depends on current definitions of what's important, which are simply inadequate for an archive. A good archivist know that it's important to keep *everything*. Archivists in the real world are, however, never very high on the pecking order. Accountants tend to rule there. As for casual, unthinking survival, often the juiciest kind, I do agree that digital material is at risk. With me dead, I suppose that the only stuff that will survive is the material that my children have. Nobody is going to keep my backups. On the other hand, my parents' material is in a number of boxes scattered around the family here and in Switzerland. I've tried finding stuff in there and frankly, it's very very difficult. A detective story each time. And of the old photos that survive, most have now been digitised and will probably survive because many members of the family now have copies- and have been able to identify many of them only because, now having copies, we have been able to email each other about them. I don't think we *know* yet how we will use 'casual' digital archives. I believe that they are potentially longer lived than physical ones. We'll see. -- Peter |