2016

[Brian Cluff]

The fix for the 32bit time problem didn't have anything to do with 
running 64bit machines since the problem is also fixed on 32 bit 
machines.  The problem is old code that specifies a 32bit number, even 
if it's running on a 64bit machine.
If any of that old code is still in use in 22 years, and we've all seen 
the articles where the airlines are still using some systems from the 
70s and such.  There will certainly still be old software being used in 
22 years....  even if it's been compiled for and running on new 128bit 
systems of the future.  Unless the time code has been updated, that 
software is going to do weird things.

I'm not saying we should panic because, like you said, it will be 22 
years from now, and the vast majority of scomputer systems will be 
replaced, but it's not something people can assume isn't going to happen 
at all.... I'm guessing a lot of people in what will be classic cars 
will be getting some nasty/funny surprises... and government computers 
are totally known to be kept completely up to date ;)

Brian Cluff

On 01/01/2016 09:04 PM, Steve Litt wrote:
> The operant word is 22 years. 22 years ago I was on a 486-66, using the
> 16 bit Windows 3.1 OS. I had just switched from floppy backups to QIC
> tape backups. I used WordPerfect 5.1 and Micrografx Windows Draw, along
> with Turbo Pascal, Turbo C, Microsoft Assembler, Clarion 2.1 for
> quickie programming, and Robohelp to make help files.
>
> Believe it or not, I still have WordPerfect 5.1 and Micrografx Windows
> Draw running on my 1998 Pentium III 300mhz running win98, but even if I
> didn't, it wouldn't be that big a deal. I can't read the QIC tapes. Big
> deal. My point is, wait long enough and today's necessities become no
> big deal.
>
> I'm pretty sure that even today's libraries, if compiled on a 64 bit
> system, will work just fine with regard to the 2038 thing. Embedded
> will be a problem, but I'm assuming that embedded for expensive stuff
> like cars will be 64 bit. Maybe your 12 year old sewing machine will
> fail, but that's about it.
>
> Meanwhile, I'd give Unix/Linux/BSD only a 50% chance of surviving the
> next 22 years. It will have to outcompete a lot of newer stuff. And if
> there's no Unix, the point is moot.
>
> SteveT
>
>
> On Fri, 1 Jan 2016 20:34:36 -0700
> Brian Cluff <brian at snaptek.com> wrote:
>
>> There's quite a few computers that are in daily use that are stuck
>> with old systems that can't be upgraded, and there are a ton of
>> embeded systems all over the place that don't get upgraded...  It
>> probably won't be a huge deal, but I'm sure something weird will
>> happen in 22 years. It could be even code running on modern systems
>> but using old libraries that use a 32bit date format.
>>
>> Brian
>>
>> On 01/01/2016 07:45 PM, Steve Litt wrote:
>>> On Fri, 1 Jan 2016 18:32:16 -0700
>>> Brian Cluff <brian at snaptek.com> wrote:
>>>
>>>> We are now almost half way, starting from y2k, to the 2038 problem
>>>> where we could see some real Y2K like problems when the older UNIX
>>>> machines run out of time its and make everything become 1901.
>>>>
>>>> https://en.wikipedia.org/wiki/Year_2038_problem
>>>>
>>>> Brian Cluff
>>>
>>> Will anyone still be using 32 bit Unix/Linux/BSD by 2038? If I'm not
>>> mistaken, on a 64 bit machine we have enough seconds to last us long
>>> beyond the probable extinction of the human species.
>>>
>>> SteveT
>>>
>>> Steve Litt
>>> November 2015 featured book: Troubleshooting Techniques
>>>        of the Successful Technologist
>>> http://www.troubleshooters.com/techniques
>>> ---------------------------------------------------
>>> PLUG-discuss mailing list - PLUG-discuss at lists.phxlinux.org
>>> To subscribe, unsubscribe, or to change your mail settings:
>>> http://lists.phxlinux.org/mailman/listinfo/plug-discuss
>>>
>>
>> ---------------------------------------------------
>> PLUG-discuss mailing list - PLUG-discuss at lists.phxlinux.org
>> To subscribe, unsubscribe, or to change your mail settings:
>> http://lists.phxlinux.org/mailman/listinfo/plug-discuss
>
>
>