Posts by Paul Crawford

No they just miss the ball & chain from transportation...

Re: What are the alternatives?

We are going to try Zoho for this. Not sure it is perfect but maybe worth looking at. Main attraction is it appears to work cross-platform and from a web browser (that can be other than chrome).

Re: Zloirock

Tsk, the second monitor is for displaying your porn preference of yesterday!

Re: Bandwidth like Bank Reserves?

As already mentioned, it comes down to cost and typical usage.

If you absolutely want a 100Mbit uncontended link you can have one, many providers will do so, but be prepared to pay at least £500/month.

Re: Well, at least i now understand why supermarkets have no loo roll...

Only if you try and use it twice...

Re: Range & Time for a FULL charge

The UK domestic supply is typically 100A max at 230V which is 23kW. Many homes have 80A or 63A fuses instead.

Most of the supply grid was built on the assumption that no domestic customer can draw full power for more than a short time, and usually not all at the same time. Rewiring and upgrading transformers on that is going to cost a lot!

Re: S-band uplink

I think the limitation is on the use of the 400kW power amplifier, not the antenna. If it were me, I would be afraid of an accidental misconfiguration or oscillating amplifier pumping out the full 400kW at low elevation even when the modem was supposedly configured for 20kW or less.

Re: No rest for the wicked

There was a big difference though, in the early days (probably up to the 70s) it was financially infeasible to tap everyone's communications so the spooks had to be selective (i.e. thousands of target, not billions) in what they looked for and recorded, and at that point a judge's warrant was perfectly in keeping with doing so.

Today the technology to hoover up all information and cross-reference it in a database, along with voice-to-text conversion, means the only practical thing standing between governments (and large corporations) and their desire for a panopticon on the population is end-to-end (i.e. per device) encryption.

Re: I think they did the right thing

A total 10 year program cost of $150B is less than the USA spends on cosmetics in 4 years. Based on $189.90 per person per year, and presumably they are using 'per person' for adults (about 75% of the 327M today) we get around $46.6B per year. From:

https://www.statista.com/statistics/304996/us-expenditure-on-cosmetics-perfume-and-bath-preparation/

That such levels of stupidity and incompetence can be displayed by a supposedly trained pilot is difficult to comprehend

I suspect if you were on a night flight, tired, and then faced with the prospect of a terminal dive you might find it harder than you expect to fully recall your training.

That is not to absolve him for the failing, just to point out that trained wet-ware is not infallible either.

Re: Unequal clothing options

Just get a kilt and boast about your Scottish roots, invented if necessary.

Re: Hardware IDs

Not convinced the same can be said about Linux though. /etc/machine-id exists, and is discouraged from being used (by sternly worded documentation), but do you trust every application you run to follow the rules?

You can use strace on any program when started to observe attempts to open that as a file, but if they are really sneaky there is probably some way to obscure the access.

What, don't you have your white port in a pint glass?

The love that dare not bark its name?

I guess the simplest option would be making the statutory warranty on electronic devices 5 years. Then the manufacturers have to bare the cost of replacement, or make it repairable, or (good grief!) actually make reliable stuff.

Sure they will bitch about it pushing up costs for consumers, but only for those who see a 1-2 year throw-away life and the resulting environmental costs as acceptable for a few quid saved initially. I doubt most of el Reg readers would see it that way...

Re: Laughable scheme

No need to worry about the small fry as the vast majority of use is Facebook and its few chums. Simply fining them real amounts (i.e. few percent of global turnover) for any shit posted will work very well.

Why? Because they depend on advertisers' payments and the gov/HMRC can take steps against companies operating in the UK. Might shore up the treasury coffers as well...

Well if it is Uber they don't seem to stop collisions...

Re: That Twitter feed

Seems to have been taken down now. Wonder why?

Dangle them in a vat of pygocentrus piraya for quicker results.

Re: Maths is hard

Thanks for explaining that, I also just looked at the reduced growth overall and though "Eh?"

But it also makes the HPE claims look even more like hair-splitting in an attempt to cover a shit decision.

Re: What will actually happen?

It really depends on the code. You might get timer loops locking up for ~67 years depending on how they are doing the time comparison / arithmetic. You might get date conversion failing (or maybe just working), again depending on how the libc code (or others) implemented the year/mon/day etc to time_t (and vice-versa) conversion.

Best case it simply works. More likely you need to reboot post-Y2038 to reset timers and then its fine. Worst case some code never works after that data. Test, test, and test gain!

Re: I've been wondering...

I think if several major companies got together and agreed on a single *new* standard that works for all then the EU would have no problem with that.

Alternatively they could mandate the companies provided free and more importantly with the product an adaptor. You would be amazed at how quick they would change "special" designs to avoid any extra production costs...

Re: Are they still around?

Sadly they are, much like gonorrhoea

Re: 2038

What seems to be missing here is the original topic - trying to fix 2038 in existing 32-bit Linux programs.

The only proper fix (i.e. works as before with signed time_t and works beyond 2038) is to go 64-bit for time_t at least (other integers can be still be 32-bit if you are targeting 32-bit MCUs etc). But that needs a re-compile of the OS and the applications (Perl interpreter in your case). And beyond that it also requires anything that maps time_t in to some structure to accommodate that. Fine for internal memory structures that are re-defined on compile to use 64-bit time_t but no use for code that has time in files, shared memory, etc, so a re-compile for 64-bit is not going to work that easily unless the program already has a debugged 64-bit version.

My point was to fix / fudge existing compiled 32-bit code the least-worst option is to make the Linux system libraries treat time_t as unsigned so it runs 1970 to 2106. Yes it breaks pre-1970 date conversion (but that was never defined as working anyway) but there are no better alternatives. Changing the epoch (even dynamically via environment variable per program, etc) will also risk breaking code that has pre-compiled time-points based on 1970, etc. If your machine is running day-to-day, and not working for very old dates, going unsigned for the internal conversion will work well enough. But there will always be something that breaks if you change it, is that any worse than everything breaking post-2038?

Well they have the stick of withdrawing Google's "essential" core code if they don't play. After all, that is the reason that an Android fork has had so little backing because (as MS found out the expensive way) no one buys the OS - they buy the phone for the apps that run on it.