|
|
Subscribe / Log in / New account

Linux support for digital video broadcasting

Mauro Carvalho Chehab, the maintainer of the kernel's media subsystem, has posted the first two in a series of articles on digital video broadcasting support in Linux. Part 1 gives an overview of how the devices and protocols work, while part 2 looks at digital TV network interface use. "Supporting embedded Digital TV hardware is complex, considering that such hardware generally has multiple components that can be rewired in runtime to dynamically change the stream pipelines and provide flexibility for things like recording a video stream, then tuning into another channel to see a different program. This article describes how the DVB pipelines are setup and the needs that should be addressed by the Linux Kernel."

to post comments

Linux support for digital video broadcasting

Posted May 30, 2015 1:20 UTC (Sat) by flussence (guest, #85566) [Link] (5 responses)

Wow, that's some good timing - I was poking around with an old WinTV card last night to see what Linux could do with it.

Seems like there's no usable signal where I am. :(

Linux support for digital video broadcasting

Posted May 30, 2015 9:39 UTC (Sat) by zdzichu (subscriber, #17118) [Link]

Most countries switched off analog TV signal already: http://en.wikipedia.org/wiki/Digital_television_transitio...

digital video broadcasting

Posted May 30, 2015 15:08 UTC (Sat) by tialaramex (subscriber, #21167) [Link] (3 responses)

In much of the world the most likely source of input today for an analogue WinTV card (they also made DVB cards under this brand name, if you have one of those it possibly should already work) is the video out of a 1980s home computer or video game intended to be plugged into a television via the "aerial socket". Broadcast using the 20th century PAL and NTSC systems is rare and found only in places with weak infrastructure and low commercial interest in radio frequency spectrum - the analogue TV systems were incredibly bandwidth inefficient.

Of course broadcasting DVB data (DVB-T via terrestrial radio or DVB-S from satellite) is also probably not going to last long. For all but the top live sporting and political events a network streaming service is cheaper to run and available to a large and growing population not limited by local geography. For a company like Netflix, broadcasting looks like a mug's game.

digital video broadcasting

Posted May 30, 2015 16:01 UTC (Sat) by mchehab (subscriber, #41156) [Link]

> In much of the world the most likely source of input today for an analogue WinTV card (they also made DVB cards under this brand name, if you have one of those it possibly should already work) is the video out of a 1980s home computer or video game intended to be plugged into a television via the "aerial socket".

Well, half of the world (e. g. most Countries in the South Emisphere) didn't stop broadcasting Analog TV yet. Also, there are still lots of Cable TV networks using analog, including hotel and in-building CATV systems like the ones used for surveillance cameras (although those are gradually being replaced by IP cameras). So, WinTV and similar cards still have its usage.

But yeah, most of the nowadays consumer's usage for those boards in US and Europe is to watch/digitalize old VCR tapes and video games.

> Of course broadcasting DVB data (DVB-T via terrestrial radio or DVB-S from satellite) is also probably not going to last long. For all but the top live sporting and political events a network streaming service is cheaper to run and available to a large and growing population not limited by local geography. For a company like Netflix, broadcasting looks like a mug's game.

I don't believe in the end of digital TV broadcasting anytime soon. TV should stay there for decades, as broadcasting has a huge penetration, specially if you count the satellite broadcasts, and require a way less bandwidth than peer-to-peer Internet connections like the ones used by Netflix and similar services. So, for me, both TV and video over IP will be in usage for a very long time.

digital video broadcasting

Posted May 30, 2015 16:55 UTC (Sat) by flussence (guest, #85566) [Link] (1 responses)

> In much of the world the most likely source of input today for an analogue WinTV card (they also made DVB cards under this brand name, if you have one of those it possibly should already work) is the video out of a 1980s home computer or video game intended to be plugged into a television via the "aerial socket".
The one I have is a multi-function card and shows up as 14f1:8800 (cx8800), 14f1:8811 (cx88-audio) and 14f1:8802 (cx88-mpeg) in lspci. There's three analogue inputs for tuner/composite/s-video, and those all work fine for that use case you mentioned, but the latency is a bit too much to replace my real TV for games (about 100ms). It also has audio RCA inputs, but I can't seem to get anything from those, and it looks like the DVB stuff works completely independently of the rest.

It's a fancy piece of hardware but like I said, no signal for me - that's down to geographical features where I live. At least it might make for a good RNG...

digital video broadcasting

Posted May 30, 2015 18:10 UTC (Sat) by Funcan (subscriber, #44209) [Link]

You can use it as an SDR receiver and pick up a bunch of stuff... broadcast FM, some HAM stuff (the space station for example), pagers (POCSAG), garage remotes, air and sea position beacons, etc

Linux support for digital video broadcasting

Posted May 30, 2015 19:29 UTC (Sat) by allesfresser (guest, #216) [Link] (68 responses)

Am I correct in thinking the DVB described in these articles is not the same thing as the ATSC broadcasting used in the United States? I had an eyeTV device a while back, but it seems there aren't a whole lot of devices available in US markets for receiving digital OTA broadcasts, at the moment.

ATSC

Posted May 30, 2015 22:02 UTC (Sat) by tialaramex (subscriber, #21167) [Link] (66 responses)

Same basic concept, almost completely incompatible technically. If you squint hard you can make out a technical argument for choosing ATSC for the specific problems the US had with digital migration - so it's not purely a NIH thing, but the result is basically the same as the US insisting on its own paper sizing, its own system of road signs and so on...

In terms of software, obviously the ATSC devices require separate drivers, which exist for some models but there is no standardisation a la webcams, so be careful or ensure you can return unexpectedly incompatible purchases. The drivers will reflect any supported device into the LinuxTV v4l2 APIs just as they would a DVB device. You will also need ATSC-specific software to sort out tuning and metadata, e.g. what channel is where, which programmes are on, that sort of thing. I believe that already exists, but I've never had to use it because I live firmly in DVB-T territory. The actual video data broadcast, in the end, is still an MPEG2 stream, so that's fine.

ATSC

Posted May 30, 2015 22:14 UTC (Sat) by dlang (guest, #313) [Link] (63 responses)

> but the result is basically the same as the US insisting on its own paper sizing, its own system of road signs and so on

to be fair, those differences developed at a time when there was very little standardization internationally. The fact that Europe has mostly thrown out their old differing standards in favor of a common set has a lot to do with how easy is was to move from an area where one set of standards prevailed to an area where a different set prevailed and the corresponding confusion. Similarly, differences from state to state withing the US have been eliminated over time and the result is two differing sets of standards, each covering a large area. I don't know how closely do China and Russia follow either standard? and if they do, how much of that is due to where they purchased equipment from? I know that Japan has a mess in their Electric Grid that boils down to the fact that different parts of the country got their grids started by buying from different, incompatible countries, and the grids grew and are bridged, but not really merged as a result.

It's _really_ expensive to switch some of this stuff, and other than the "but now we're standard", there isn't a lot of benefit to those who have to suffer the cost of switching. International companies that can then standardize benefit from the simplification, but it takes a long time before the man in the street who has to replace a lot of stuff to match the new standards will see any benefit.

ATSC

Posted May 31, 2015 10:29 UTC (Sun) by tialaramex (subscriber, #21167) [Link] (62 responses)

For paper sizes, the situation is _really_ simple.

Europe didn't just coalesce onto one arbitrary standard, the "best man won". The Germans (isn't it always the Germans?) had sat down and designed a completely rational self-consistent system by the 1920s‡. By the 1970s that system was widespread in Europe and went to ISO where it became ISO 216. Remember when ISO standard numbers were only three digits long? Me neither. The A-series paper system is so much more sensible than the average "It came out of my paper press this size, what more do you want?" previous approach that it took over the entire world, except the US and her immediate neighbours.

‡ Here's how this goes. Find out the square root of 2. Rectangles in this ratio have a convenient property, if you fold them in half on the long axis you get another rectangle with the same ratio but half as big. Amazing! OK, now make a rectangular piece of paper in this ratio but with an area of 1 metre squared. There is only one possible way to do this. This is called A0 ("A-zero"). If you fold it in half, as described previously, that size is A1, and so on, A2, A3, A4, all the same ratio but smaller. No need to buy special paper to print an eight page pamphlet in A5, you can print it on A4 and fold it. Designed an A5 front page but now want to use it for an A3 poster? No problem, just scale it up 'cos they're the same ratio. If you have the metric system (again everybody but the US) then paper density calculations also become trivial. If you're using 80gsm paper then the mass of 1 sheet of A0 in that paper would be 80 grams, because A0 has a metre squared of area, thus an A1 sheet has mass of 40 grams, A2 is 20 grams, A3 is 10 grams, A4 is 5 grams. 27 sheets of A4 will be 135 grams. Rational!

You wondered about the expense. The US Federal government (inevitably) performed a study. The study said it should switch, because even internally the cost saving would be significant over time. The present irrational non-self-consistent system causes endless trouble that people who've grown up with it just accept as normal, that would go away. However there is of course zero political support for switching, Americans may hate government waste but they _love_ patriotism. "My country, right or wrong, but especially when wrong".

ATSC

Posted May 31, 2015 11:03 UTC (Sun) by paulj (subscriber, #341) [Link] (48 responses)

And the worst part is that lots of software defaults to these inconsistent, ad-hoc US paper sizes.

Ditto for bloody imperial - which isn't consistent between England and USA (gallon? Uh, which gallon?). Even the UK uses metric for most stuff these days, so only a tiny minority of the world uses non-metric by default, yet the vast majority have to be afflicted with imperial units as the default and figuring out how to change them.

If it is even *possible* to set a default.

Some software doesn't let you specify a default, but instead sets your units according to geo-location. So if the software decides you're in the UK, you start getting distances and speeds based in idiotic miles and feet (thank you Google Maps). If in the USA, even worse, you might get the whole gamut of bizarre measures from the days when most people were innumerate.

Sigh.

Dear US software developers developing global products: Please realise you're in a tiny minority when it comes to your use of antiquated, inconsistent measures.

ATSC

Posted May 31, 2015 19:59 UTC (Sun) by dlang (guest, #313) [Link] (18 responses)

> Dear US software developers developing global products: Please realise you're in a tiny minority when it comes to your use of antiquated, inconsistent measures.

well, developers write for themselves and people near them.

If they are usually defaulting to US measurements, that says something about the proportion of developers in various places, and probably says something about the proportion of users of their software.

If the US really is "a tiny minority" of users of the software, you would expect that the developers would be getting enough complaints to change the default. If however, the US is not a tiny minority of the software users (I assume of software written for English users), then your complaint boils down to wishful thinking.

Further divergence from the topic

Posted May 31, 2015 20:36 UTC (Sun) by tialaramex (subscriber, #21167) [Link] (15 responses)

"you would expect that the developers would be getting enough complaints to change the default"

No. In order to get a complaint the person complaining has to understand that things needn't be this way (obvious enough for the LWN.net readers but not for say, my sister) and care so much they jump through hoops to moan about it. Unless the program is used largely for printing, and maybe there is no way to permanently change the default, there's no way the handful of resulting complaints would percolate to the top of the pile.

On most operating systems today the information about locale (how should we display dates? What size of paper do people use here? How should I sort names of things?) is provided by the OS itself, not something every program should insist on determining for itself. Thus this is a QOI issue. Chances are that programs not paying this element of their "taxes" are missing out other things too. Maybe they don't work properly on multi-monitor setups, they can't do IPv6, they assume the user's profile is stored on a local disk with high I/O bandwidth, they expect the network to be available even when the user isn't trying to use it, they only work with a relative pointing device not an absolute one...

Further divergence from the topic

Posted May 31, 2015 21:35 UTC (Sun) by butlerm (subscriber, #13312) [Link] (8 responses)

Speaking of international standards, at the rate we are going IPv6 is going to be one of the greatest wastes of resources in the history of software development. Not that people shouldn't make reasonable accommodations for people forced to use such an ill fated network protocol, but in the long run it will probably all be for naught.

Further divergence from the topic

Posted Jun 1, 2015 11:01 UTC (Mon) by jem (subscriber, #24231) [Link] (5 responses)

IPv6 is doing fine. Google is measuring the percentage of users accessing Google over IPv6, and publishes statistics over here: https://www.google.com/intl/en/ipv6/statistics.html

Just this weekend the global figure passed 7 %, and the figure for the US is 18 %; 33 % for Belgium. I suspect a big portion of these are mobile users and the operators provide a NAT64 service to be able to access IPv4-only sites.

Further divergence from the topic

Posted Jun 1, 2015 19:35 UTC (Mon) by HIGHGuY (subscriber, #62277) [Link] (4 responses)

>33 % for Belgium. I suspect a big portion of these are mobile users and the operators provide a NAT64 service
> to be able to access IPv4-only sites.

Actually, no.
Telenet one of the major providers in Belgium supports native IPv6 and made the switch some time ago. That's why IPv6 penetration got a lot higher.

Further divergence from the topic

Posted Jun 2, 2015 12:04 UTC (Tue) by jem (subscriber, #24231) [Link] (1 responses)

Wow. Finally somebody has the guts to provide a solution that points to the future instead of investing in the dead end that NAT44 is.

I found the following presentation on the net - thanks for the tip.

http://www.ipv6council.be/IMG/pdf/20141212-05_Thienpondt_...

Further divergence from the topic

Posted Jun 2, 2015 12:37 UTC (Tue) by cortana (subscriber, #24596) [Link]

OTOH it's depressing that so much effort has to be put in to work around providers who can't be bothered to upgrade their services to operate over IPv6. You would think Microsoft of all companies would be able to afford it.

Further divergence from the topic

Posted Jun 5, 2015 11:23 UTC (Fri) by robbe (guest, #16131) [Link] (1 responses)

NAT64 usually means that the hosts (mobiles in this case) are IPv6 only. So you're in violent agreement with the OP.

(NAT64 just gives you a tool to reach the long tail of destinations that are still v4-only.)

Further divergence from the topic

Posted Jun 5, 2015 20:54 UTC (Fri) by HIGHGuY (subscriber, #62277) [Link]

Telenet is a virtual mobile operator too, but that's not the use-case.
Their docsis network has a dual-stack implementation. No NAT64 or other trickery involved...

Further divergence from the topic

Posted Jun 1, 2015 11:07 UTC (Mon) by cortana (subscriber, #24596) [Link] (1 responses)

So what is the future of the Internet, then? Carrier grade NAT, stamping on a human face, forever?

Further divergence from the topic

Posted Jun 1, 2015 16:26 UTC (Mon) by raven667 (subscriber, #5198) [Link]

The benefit of a CGN-based Internet is that it forces more and more centralization for clients and servers to be able to communicate with one another which is great if you are a gatekeeper building a bigger empire, not so great if you take a more democratic approach to services and don't want to require authorization from a service provider that you have no leverage over to enforce accountability.

Further divergence from the topic

Posted Jun 1, 2015 1:12 UTC (Mon) by dlang (guest, #313) [Link]

> Thus this is a QOI issue. Chances are that programs not paying this element of their "taxes" are missing out other things too.

I disagree with the idea that every piece of software written needs to do all these things. The software needs to do what it's users are using it for and the developers have a limited budget of time to work on things.

It's one thing for big projects (Google scale for example) to take the attitude that everything must include all of the complications needed to support all the variations used around the world. It's another to say that every program developed by volunteers must do the same thing. If the program becomes popular enough, it will get these sorts of features (or inherit them from a library that they use that is popular enough to have implemented these featuers), but if not it doesn't mean that the program is of poor quality, it just means that there are places it doesn't work as well. That's a limitation on it's Scope, not a reflection of how well it works within that scope.

Further divergence from the topic

Posted Jun 1, 2015 10:03 UTC (Mon) by paulj (subscriber, #341) [Link] (4 responses)

"On most operating systems today the information about locale (how should we display dates? What size of paper do people use here? How should I sort names of things?) is provided by the OS itself,"

If you happen to be in the UK or USA but want non-"daft units from the age of innumeracy", you may have no choice. E.g. Android only gives you a language setting, but no further locale configuration settings. Even if your desired language corresponds to your choice of units, or your OS gives you decent configuration options, some apps may choose to ignore it - e.g. Google Maps in my browser on my desktop seems to go according to geo-ip. I can change to kilometres in the options, but it doesn't remember it.

Further divergence from the topic

Posted Jun 1, 2015 14:00 UTC (Mon) by rschroev (subscriber, #4164) [Link] (2 responses)

> Google Maps in my browser on my desktop seems to go according to geo-ip.

Google Maps seems to select not based on my location, but based on the map location. Driving directions here in Belgium from e.g. to Antwerp to Brussels use kilometers, but directions from Washington DC to New York City use miles.

I guess there's some logic in that behavior (signs along the route will also use kilometers respectively miles), but it's not very useful to me.

Further divergence from the topic

Posted Jun 1, 2015 19:14 UTC (Mon) by cebewee (guest, #94775) [Link] (1 responses)

This behaviour can be changed: If you ask maps for directions, there is a settings button (route options, with a few sliders). Here you can set your preferred units. I had to search a few minutes to find this settings, even though I knew it existed ...

Further divergence from the topic

Posted Jun 1, 2015 22:05 UTC (Mon) by paulj (subscriber, #341) [Link]

There was an option, but it didn't "stick". You had to set it every time.

Now, my Google maps has changed its UI quite noticably just this week, so they're possibly rolling out some update, and I don't see any option anymore. I'm stuck with moronic miles.

Further divergence from the topic

Posted Jun 5, 2015 12:14 UTC (Fri) by robbe (guest, #16131) [Link]

> If you happen to be in the UK or USA but want non-"daft
> units from the age of innumeracy", you may have no choice.

On a libre system, you have choice. If all else fails, change the right source file (e.g. http://sources.debian.net/src/glibc/2.19-18/localedata/lo...) and rebuild.

Most Android-based setups and Google Maps are not libre, of course.

ATSC

Posted Jun 1, 2015 9:34 UTC (Mon) by paulj (subscriber, #341) [Link]

"well, developers write for themselves and people near them."

Hence why I qualified my call to US developers with "developing global products", i.e. those who (if they thought about it) desire a global audience for their software.

ATSC

Posted Jun 5, 2015 12:02 UTC (Fri) by robbe (guest, #16131) [Link]

> well, developers write for themselves and people near them.
Sure, and why should I care about my software working on a non-systemd, or non-Ubuntu, or non-OSX system, when I use a systemd, or Ubuntu, or OSX system ...

> If the US really is "a tiny minority" of users of the software,
> you would expect that the developers would be getting
> enough complaints to change the default.
I don't know how many reports developers hear about that. But I can imagine a couple of other factors that would keep you from complaining about such an issue.
* you don't think that your opinion is important
* you don't know English all that well
* you don't think that it's such a big issue

ATSC

Posted Jun 1, 2015 10:23 UTC (Mon) by Seegras (guest, #20463) [Link] (26 responses)

> Ditto for bloody imperial

Funny enough, most republicans in the US are staunch supporters of the imperial system, and not the republican (from the French Republic, the metric) one. Does this want to tell us something? :)

ATSC

Posted Jun 1, 2015 12:33 UTC (Mon) by anselm (subscriber, #2796) [Link] (25 responses)

US liquid gallons are not “imperial” – a US liquid gallon is not quite 3.8 litres but a (UK) “imperial” gallon is about 4.5 litres.

ATSC

Posted Jun 1, 2015 12:43 UTC (Mon) by paulj (subscriber, #341) [Link]

If only there were some consistent, coherent system of units of measure that were internationally recognised and used by 95%+ of the planet.

"Imperial" measures

Posted Jun 2, 2015 9:11 UTC (Tue) by farnz (subscriber, #17727) [Link] (23 responses)

Both gallons start life as British measures, however. Historically, Britain used different volume measures for different liquids; at the time the US rebels broke away from the British Empire, we still had (at least) the wine gallon (around 3.8 litres, believed to have been chosen as it was 10 troy ounces in weight of wine) and the ale gallon (around 4.5 litres, believed to have been chosen as 10 avoirdupois ounces of beer). As time went on, the US settled on the wine gallon as the standard gallon, while the Empire settled on the ale gallon as the standard gallon.

Perhaps there'd be more pressure to go to metric measures if people still had to deal with different units depending on what you're measuring - having to know whether the "gallon" in question is an ale gallon, a coal gallon, a corn gallon, or a wine gallon, and whether you need to convert between the various gallons for this use case might just kick people into moving to the global system.

"Imperial" measures

Posted Jun 8, 2015 18:19 UTC (Mon) by Wol (subscriber, #4433) [Link] (1 responses)

Actually, the measure which is "off" is the pint ...

One American oz == One British oz

One American gallon == 8 American pints

One British gallon == 8 British pints

One American pint == 1 fluid pound == 16 fluid oz

One British pint == 20 fluid oz

I have no idea why we settled on 20 fl oz as our pint.

Cheers,
Wol

"Imperial" measures

Posted Jun 8, 2015 21:03 UTC (Mon) by mjg59 (subscriber, #23239) [Link]

> One American oz == One British oz

The US and UK fluid ounces are similar, but not the same.

"Imperial" measures

Posted Jun 8, 2015 18:22 UTC (Mon) by Wol (subscriber, #4433) [Link] (20 responses)

> Perhaps there'd be more pressure to go to metric measures if people still had to deal with different units depending on what you're measuring

Except the metric system is just as bad ... how do I know whether I'm getting value for money when I buy a litre of ice cream?

(The manufacturers like to stuff ice cream full of air, so when buying cheap ice cream you're probably paying over the top :-(

Cheers,
Wol

"Imperial" measures

Posted Jun 8, 2015 19:41 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (15 responses)

So buy half a kilogram of ice cream. Duh.

All European countries mandate labeling that includes net weight.

"Imperial" measures

Posted Jun 8, 2015 20:36 UTC (Mon) by Wol (subscriber, #4433) [Link] (14 responses)

I don't think I've ever seen (in the UK) a tub of ice-cream with the weight on it !!! And I *have* looked !!!

I do know the Co-op ran into serious legal difficulties trying to put a "laid on" field on their eggs, (in addition to the legally mandated "best before"). As I understand things, the legal measure for ice-cream is volume, therefore if a manufacturer wants to put the weight on it they will have problems ...

In the UK certainly, what with the attempt to stamp out the usage of the pound, there were real problems labelling foodstuffs with anything more than the legal *minimum* requirement - yes that's right, it was *illegal* to provide *extra* information for consumers!!!

(NB - I have just examined a tub of ice cream in my freezer. The ONLY quantity information it gives me is the volume - two litres. So labelling by weight is clearly NOT mandatory and, I suspect, in the UK it's actually illegal :-(

Cheers,
Wol

"Imperial" measures

Posted Jun 9, 2015 0:12 UTC (Tue) by cesarb (subscriber, #6266) [Link] (1 responses)

I wonder how it's in my country (Brazil).

<one quick Google search later>

Since I didn't have any ice cream at home right now, I searched for one of the most popular ice cream brands, and randomly clicked one of the top image search results.

I clearly see in the (blurry) photo I chose the label on top of the box: PESO LÍQ.: 1,021kg / CONTEÚDO: 2L.

Not an European country, but interesting anyways.

(...weren't we supposed to be talking about television?)

"Imperial" measures

Posted Jun 9, 2015 11:02 UTC (Tue) by adobriyan (subscriber, #30858) [Link]

Just bough an ice cream ("plombir"-something) in small shop in the "office" building (former soviet glory) in Belarus.

Netto mass is clearly printed in bold font, the same font is used for the name of ice cream itself.

But the "95g" disease is there. :-(

"Imperial" measures

Posted Jun 9, 2015 9:37 UTC (Tue) by paulj (subscriber, #341) [Link] (11 responses)

That's not really an issue with the metric system though, is it? :) That's to do with food regulation requiring volume and (from what say) disallowing weight units. The underlying system of measures is irrelevant?

"Imperial" measures

Posted Jun 9, 2015 10:18 UTC (Tue) by Wol (subscriber, #4433) [Link]

No - but I was responding to this comment

> Perhaps there'd be more pressure to go to metric measures if people still had to deal with different units depending on what you're measuring

The problem being that if you measure stuff of VARIABLE density by volume, the resulting measurement is pretty much meaningless. Unfortunately, this is very common in the Anglo-Saxon world, the American propensity to measure various types of powder in "cups" is just as bad. I know a cup is a standard volume, but is what you're measuring a standard density?

Where's the pressure to go metric, if the "different units" aren't actually appropriate for purpose?

Cheers,
Wol

"Imperial" measures

Posted Jun 9, 2015 10:25 UTC (Tue) by Wol (subscriber, #4433) [Link] (9 responses)

> and (from what say) disallowing weight units.

Not sure when, it was the John Major government, so it was about 15 years ago at least ...

They force-metricated the weights by which food was sold, so you had to advertise the weight in kilos or grams. We still buy stuff in pounds, it's just 484g (iirc). They haven't metricated the quantity and rounded it to 500g.

There were a couple of extremely public cases (the authorities threatening to take traders to court) because said traders kept on displaying weights in pounds. Allegedly (and probably truthfully) because they had a load of older customers who just couldn't get to grips with kilos. After all, I still sometimes think in LSD, and when did our currency go decimal? 1971 !!! That really is my biggest gripe with the EU - there are too many idealists who know where they want us to go (and usually I agree with them), but they are determined to FORCE us there AS QUICKLY AS POSSIBLE, and people just aren't ready for it! It's a sad truism, but if you want change your best bet is to change it for the young, and just let the older people die off ... :-)

Cheers,
Wol

"Imperial" measures

Posted Jun 9, 2015 12:26 UTC (Tue) by micka (subscriber, #38720) [Link] (6 responses)

> After all, I still sometimes think in LSD

in -> under
And yes, I think it doesn't help.

"Imperial" measures

Posted Jun 9, 2015 15:33 UTC (Tue) by paulj (subscriber, #341) [Link] (5 responses)

He means the old, batshit insane system of money denominations the UK had up until 1971: http://en.wikipedia.org/wiki/%C2%A3sd

LSD = pounds (L/£), shillings (s), pence (d). In addition they also had sovereigns, crowns, thrupennies (3d), ha'pennies (½d), sixpenny bit (6d) - and no doubt more. There were 5 pennies in a shilling, 240 in a pound. Absolutely bonkers, and making not a shred of sense to anyone with the most basic grasp of doing calculations in the dominant number system for the last thousand+ years (which, granted, may been a very limited number of people until not so long ago).

"Imperial" measures

Posted Jun 9, 2015 15:46 UTC (Tue) by mgb (guest, #3226) [Link]

240 pennies (960 farthings) to the pound was good for the same reason that 360 degrees per circle is good. Similarly a shilling was 12 pennies or 48 farthings and also divisible in lots of useful ways. Decimal suffers from a serious lack of whole number divisors.

"Imperial" measures

Posted Jun 9, 2015 18:01 UTC (Tue) by Wol (subscriber, #4433) [Link]

> There were 5 pennies in a shilling

You've got your old and new currencies mixed up - it's 5 new pence to one old shilling.

And while yes we did have all those weird denominations, it's no worse than the American nickel/dime/penny which totally confuses foreigners to whom those names mean absolutely nothing in terms of dollars and cents. We all knew the assorted "conversion rates" from crowns, tanners, bobs, florins etc to their lsd equivalent. And most if not all of those were actually the names of coins, so it all made a lot of sense :-) Even a half-crown.

Oh - and you've forgotten groats, which feature a lot in literature. I'm guessing they disappeared at the end of the Victorian era - I'm too young to remember farthings (just) - they got rid of the ha'penny when I was a youngster. I *think* the groat was 4d, but it could have been two groats to a farthing.

And don't forget, until inflation went mad in the 70's most peoples' income was measured in shillings, so 12 pence to a shilling wasn't that difficult to cope with ... same as 12 eggs to a dozen, or 12 inches to a foot ...

Cheers,
Wol

"Imperial" measures

Posted Jun 9, 2015 22:34 UTC (Tue) by nix (subscriber, #2304) [Link] (2 responses)

While it *is* bonkers, the Wikipedia article you link to actually provides some rationales (and yes, it's a bit more than a thousand years since Charlemagne reintroduced it -- it originated in the Roman Empire). If things are generally provided in dozens -- and note how many ancient staples like eggs *are* -- then the conversions generally cancel out, so you can go straight from 'a dozen eggs cost two shillings' to 'an egg costs twopence'. Everything has lots of factors, just as with the other parts of the Imperial system, which might possibly make mental arithmetic easier iff you used those systems all the time.

(Thank goodness I don't have to. Metric everywhere!)

"Imperial" measures

Posted Jun 10, 2015 11:19 UTC (Wed) by paulj (subscriber, #341) [Link] (1 responses)

Oh yes, shilling and penny clearly have much older roots. I'm aware.

I'm also aware 12 was a very common base for things in times past. However, duodecimal bases don't really make much sense anymore when your society has shifted to a decimal number system many hundreds of years ago. I assume widespread innumeracy is why that was tolerated. People just "knew" how many Xes there were to a Y so...

That 3 and 4 make up better factors for collections of things as an argument against metric or decimal is also just bizarre. Nothing in metric or decimal stops you from having collections of things that are products of 3 and 4. The fact you can describe a dozen eggs in decimal shows this. Similarly, you can state there are 240 pennies.

What doesn't make sense are ad-hoc multiplicative divisions that simply make *no sense in any number system*. E.g. 240 pennies in the old pound? There is no power of 3 or 6 with which it aligns. It's an order of divisions which makes it impossible to construct regular arithmetic operations on.

If there had been 6 pennies in a shilling, 6 shillings in a florin, 6 florins in a crown, and 6 crowns to a pound, maybe it'd have been defensible. However, there weren't. It was a bonkers, irregular system that made little sense, was not amenable to calculations, and could only have been "designed" and used by people with a limited understanding of numbers.

"Imperial" measures

Posted Jun 16, 2015 15:20 UTC (Tue) by nix (subscriber, #2304) [Link]

E.g. 240 pennies in the old pound? There is no power of 3 or 6 with which it aligns. It's an order of divisions which makes it impossible to construct regular arithmetic operations on.
That was clearly for groups of groups of things: the first group goes from pennies to some number of shillings, the second from shillings to some number of pounds. Arguably embedding this sort of thing in the currency is nuts, but that's how evolved systems tend to end up...

"Imperial" measures

Posted Jun 9, 2015 22:30 UTC (Tue) by nix (subscriber, #2304) [Link] (1 responses)

This was a classic example of the UK being the country in the EU that most overenthusiastically enforces EU measures, way past their actual wording. There is no EU directive of any sort prohibiting putting as many measures as you like on things -- one of them just has to be metric, is all. But a bunch of power-mad local authority bureaucrats thought that making everything just one unit would make enforcement easier or something (I don't think anyone has ever successfully teased out the *reason* for this).

Of course, certain newspapers immediately blamed the EU (as they blame the EU for everything up to and including old age and death) -- but the EU cannot be blamed for this one.

"Imperial" measures

Posted Jun 10, 2015 16:06 UTC (Wed) by Wol (subscriber, #4433) [Link]

:-) too true ... :-(

But I think there is a lot of pressure from Europe for a "United States of Europe" and they charge in without thinking. Then you get disasters like Black Monday and the looming Grexit in the financial realm, and stupid things like the Czech government suing Britain for not paying social security to Czech nationals - hang on - we wouldn't pay social security to *British* nationals under those circumstances!!! And Schengen - I'd love us to join, but not if open borders means a flood of migrants claiming social security.

There's too much idealism, not enough pragmatism, and a massive democratic deficit. And then, as you say, we have our own bureaucrats who gold-plate EU directives at every opportunity :-(

Give me an EU parliament like the way I understand the Canadian one functions. MUCH more power (it is an elected body after all), but I gather states are not obliged to accept Federal legislation. They can't repeal it once it's taken effect in the state, but they can pass legislation that defers its implementation. So it takes a while for things to work their way through the system, but the Federal Government can't be too progressive too quick, and the states can't be seen to be dragging their heels too much.

Cheers,
Wol

"Imperial" measures

Posted Jun 9, 2015 11:57 UTC (Tue) by tao (subscriber, #17563) [Link] (3 responses)

*Of course* it's stuffed with air. The whole point of ice cream is that you're aerating a custard.
You're right, weight should definitely be listed along with volume (and I really wish that would go for
ingredients in recipes too).

As a foodie, I'm more concerned with taste though, generally knowing if they've used real vanilla rather than
vanillin says a lot more about the quality of the ice cream; same goes for whether or not it contains real eggs,
etc.

"Imperial" measures

Posted Jun 10, 2015 15:56 UTC (Wed) by Wol (subscriber, #4433) [Link] (2 responses)

Actually, as far as I'm aware ice cream is NOT an aerated custard - that's a gelato. And I believe they definitely are different, though I'm not that sure of what's what. The name makes the difference clear, I believe.

"lato" is from the italian work for milk.
"cream" is, well, cream.

So you add egg to milk to get your custard for gelato. You start with cream so you don't need egg to thicken it for ice cream.

(It wouldn't surprise me if the makers of cheap ice cream realised that gelato used cheaper ingredients, so they made a gelato and called it ice cream ... to me at least "gelato" is a new immigrant to the English language. And I don't mean the American language ... !)

Cheers,
Wol

"Imperial" measures

Posted Jun 10, 2015 17:56 UTC (Wed) by corbet (editor, #1) [Link]

"Gelato" is the past tense of the Italian verb "gelare" - "frozen". Milk ("latte") doesn't play into it (the word, that is).

"Imperial" measures

Posted Jun 11, 2015 10:01 UTC (Thu) by tao (subscriber, #17563) [Link]

I guess it's all a question of nomenclature. When I refer to something as ice cream it's generally (though not always) an aerated custard;
gelato is also aerated, just a bit less than the ice creams you'd typically find in a store. Gelato also (typically) has a lower dairy and fat content,
though the fat content varies quite a bit based on what thing you flavour it with.

Some people even call sorbet & granita ice cream; I certainly don't, seeing as they contain no cream -- or other suitable replacements -- at all.

All that said, I've by now made 40-50 different kinds of ice cream, sorbet, granita, gelato, etc. Having a proper (with compressor)
ice cream maker (for ice cream, sorbet & gelato) and a high power mixer that can crush ice (for a poor man's granita when I don't have the patience)
makes for a lot of fun.

Imperial Measurements

Posted Jun 4, 2015 23:05 UTC (Thu) by kenmoffat (subscriber, #4807) [Link] (1 responses)

Here in the UK, although we still measure long distances in miles (I did once see a road sign which also had a km measurement, but only once ;) and we still use miles-per-gallon for cars, but many, or event most, people use Celsius for degrees (although we probably call it Centigrade). I recently upgraded my mobile phone - the new one runs android 4.4, but the google app for news insists on showing me the temperature here in Fahrenheit - for me, that is totally meaningless (I can remember that 32F is 0°C, but not much else) but there is apparently no way to configure it so that part is a total waste of space. Fortunately, I have the BBC app for checking the weather.

Imperial Measurements

Posted Jun 5, 2015 20:13 UTC (Fri) by cesarb (subscriber, #6266) [Link]

> I recently upgraded my mobile phone - the new one runs android 4.4, but the google app for news insists on showing me the temperature here in Fahrenheit [...] but there is apparently no way to configure it

If it's the same Google app for news that's installed on mine (since mine's was updated through Google Play, it should be the same for all devices with up-to-date Google apps), there's a setting for it. Drag from the left border to show a hidden menu, then scroll to the bottom where there should be a Settings entry. The last two settings are temperature units and wind speed units.

Paper Size

Posted May 31, 2015 19:18 UTC (Sun) by tonyblackwell (guest, #43641) [Link]

You taught me something I didn't know. Fascinating. Thanks.

ATSC

Posted May 31, 2015 20:06 UTC (Sun) by dlang (guest, #313) [Link] (10 responses)

I can see where an International company would benefit from one standard paper size around the world. But unless I am developing posters/flyers where I want the same thing from a handbill to a big poster (which I think is unlikely in any case), where is the benefit to anyone else?

the 'ease of calculating the size or the weight' don't ring true to me. Normal people don't calculate such things, they either measure them or look them up.

If a business changes the paper size they use, they will end up with a significant period of inconsistency where paper from two different places or times are of slightly different sizes. What's the benefit of going through all this pain?

and no, the US paper sizes are not "It came out of my paper press this size, what more do you want?" they are a standard (admittedly not as nicely consistent or mathamatically nice as the ISO standard)

remember that the nice thing about standards is that there are so many to choose from :-)

ATSC

Posted Jun 1, 2015 11:03 UTC (Mon) by geert (subscriber, #98403) [Link]

Normal use cases:
- What envelope (yes, this extends to envelopes, too!) do I need to send a bunch of A4 pages? A C4 envelope. For an A4 page folded in two? A C5 envelope.
- Which stamp do I need to send an envelope containing 3 A4 pages? (3 + 1 (for the envelope)) pages weigh ca. 4 x 5 grams.

paper

Posted Jun 1, 2015 12:54 UTC (Mon) by tialaramex (subscriber, #21167) [Link] (8 responses)

> and no, the US paper sizes are not "It came out of my paper press this size, what more do you want?" they are a standard

This is true in a sense, ANSI wrote a document in 1995 that formally standardised the crazy US sizes, so if you need a citation for them you can quote that 1995 ANSI standard today.

But hopefully you don't believe US Letter sprang into existence in 1995. So, where does it originally come from? We aren't sure. The US government did repeatedly attempt to standardise paper sizes, but none of the proposed sizes match US Letter until the late 20th century by which time we know it was in widespread use outside the government. A common "raw" paper size used to estimate "basis weight" (the crazy American way to specify paper densities) can be cut down to either US Legal or US Letter, which is the only reason "basis weight" for both types of paper is the same for the same density of paper. But the most likely origin of /that/ raw size is er, that's the size it comes out of the mould in artisan paper production. Which you rejected as the reason, so obviously you've got an alternative?

paper

Posted Jun 1, 2015 17:14 UTC (Mon) by dlang (guest, #313) [Link] (7 responses)

"that is the size that happened to come out of the press" works for one paper producer, it doesn't explain why any others opted to use the same size.

Paper sizes had to standardize for the printing press to succeed and paper hit mass production. It doesn't work if your paper doesn't match the size of the press. If the paper production is local to the press, both can be an odd size, but if the paper manufacturer is going to sell paper to more remote people, they are going to need to use the same size.

paper

Posted Jun 1, 2015 18:57 UTC (Mon) by PaXTeam (guest, #24616) [Link] (6 responses)

> It doesn't work if your paper doesn't match the size of the press.

thank god we have also invented blades that can cut through the mess ;). btw, instead of speculating on how the press works, maybe pay a visit to a real one (and also see how much waste they generate in a day, on standard paper sizes).

paper

Posted Jun 1, 2015 23:17 UTC (Mon) by anselm (subscriber, #2796) [Link] (5 responses)

The A series of paper formats is convenient in many ways but does not really lead to good sizes for books. It turns out that if you want to use, say, A5-sized pages for something like a novel (usually by cleverly arranging “signatures” of, e.g., 16 pages' worth on one sheet of A1 paper so that when the sheet is folded down to A5 size and stacked with the rest of the book, everything ends up where it belongs and you just need to trim the edges a little to make the stack of pages open in the right places), you either end up with very wide margins or else with lines that are too long to read comfortably and/or a book that is inconvenient and tiring to hold up. So if you're a book printer, you pick a size that you like and cut off whatever is too much such that the end result is reasonable. (The situation is not that different with the US paper sizes.)

This of course does not detract from the fact that US paper sizes suck and the US should convert over to what the rest of the world is using ASAP.

paper

Posted Jun 2, 2015 0:28 UTC (Tue) by dlang (guest, #313) [Link] (4 responses)

so what is the advantage of the different paper size other than the fact that it's dimensions are nice round numbers in metric?

paper

Posted Jun 2, 2015 0:52 UTC (Tue) by mjg59 (subscriber, #23239) [Link]

As mentioned earlier, you benefit from ease of scaling (up or down - scale A4 down by 50% and you can fit two pages to every piece of paper without additional empty space or altering aspect ratio, scale A4 up by 100% and you can turn what came out of your home printer into something poster sized), ease of weight calculation (A0 is a metre squared, weight is in grammes per square metre), it being easier to fold larger paper into smaller envelopes (want to put A3 in a C4 envelope? Fold it in half so it's A4, go from there), it being possible to fold large documents into standard-sized folders (Keep folding A0 in half and it'll drop into an A4 file), and probably others.

It's paper. It's not going to cure hunger. It's not going to bring about world peace. The advantages are clear, but you can easily argue that the they aren't worth the drawbacks of transition. I've certainly got no idea whether the economic cost would be worth it. But since you've apparently put sufficiently little research into the topic that you didn't even notice that the dimensions *aren't* nice round numbers in metric, I don't think you have either.

paper

Posted Jun 8, 2015 20:41 UTC (Mon) by Wol (subscriber, #4433) [Link] (2 responses)

> other than the fact that it's dimensions are nice round numbers in metric?

The fact that a book is a convenient size to hold easily.

The fact that a book is a convenient width to read easily.

Ergonomic factors come seriously into play. My wife has great difficulty reading a typical A4 sheet of paper, she can no longer keep track of which line she's on because as she jumps from the end of one line to the beginning of the next, she loses her place. This is very common as you get older.

That's why newspapers have columns - people couldn't read them otherwise.

Cheers,
Wol

paper

Posted Jun 8, 2015 20:59 UTC (Mon) by dlang (guest, #313) [Link]

umm, how does A4 vs Letter size have that significant a difference?

The statement was that it would be a huge benefit to have the US switch from letter size to metric A4 size paper. I'm asking where the benefit is for anyone who's not a Government or International company.

paper

Posted Jun 9, 2015 9:38 UTC (Tue) by paulj (subscriber, #341) [Link]

How is that a function of (small) differences in paper size, rather than how the text is printed on the paper?? The latter surely.

Present irrational non-self-consistent system

Posted Jun 2, 2015 13:41 UTC (Tue) by pr1268 (guest, #24648) [Link]

The present irrational non-self-consistent system causes endless trouble

Perhaps you're confused: 8.5 and 11 (and also 14 for legal-size) are rational numbers, but √2 is irrational. Plus, to find the dimensions of 1 m2 of paper (A0), you have to take the square root of (width2 ⨯ √2)—that's right, a nested square root! The irrational just got even more irrational.

I'm only kidding about your confusion; just being a little facetious here. But my math dilemma is real. (At least it isn't imaginary!) <RIMSHOT>

I sincerely wish the USA would adopt the Metric System fully (and kick the imperial units to the curb). And yes, I'm a patriotic American. I'm just tired of seeing the US be the laughing stock of the rest of the World in this regard.

And finally, thank you for enlightening me on A-series paper sizes. Good stuff!

ATSC

Posted May 30, 2015 23:59 UTC (Sat) by Trelane (subscriber, #56877) [Link] (1 responses)

> If you squint hard you can make out a technical argument for choosing ATSC for the specific problems the US had with digital migration - so it's not purely a NIH thing, but the result is basically the same as the US insisting on its own paper sizing, its own system of road signs and so on...

The technical reason is quite valid: DVB didn't exist. Indeed, many at the time didn't believe real HDTV broadcast was realisitically feasible (iirc, it was believed to require much too much bandwidth). How it came to happen in the US (let alone replace by fiat the analog system) is a great read: http://www.amazon.com/Defining-Vision-Broadcasters-Govern...

A friend and fellow ham told me about the book, and he was correct. It's a fascinating story. I'd highly recommend that you read it. And also learn to hate on others a little less maybe. :)

ATSC

Posted Sep 14, 2015 14:40 UTC (Mon) by quboid (subscriber, #54017) [Link]

"A friend and fellow ham told me about the book, and he was correct. It's a fascinating story."

...and for the European side of the story there's http://www.amazon.co.uk/Inventing-Digital-Television-Tech... which I found to be a good read.

Linux support for digital video broadcasting

Posted Jun 1, 2015 9:36 UTC (Mon) by linuxrocks123 (subscriber, #34648) [Link]

It wasn't at all hard for me to find a card that supported ATSC. Back in 2012, I paid about $6 for a PCI capture card on eBay. Though the technologies are different, ATSC still uses the therefore confusingly named "DVB driver subsystem".

Note that some manufacturers don't advertise ATSC compatibility of their cards even when the internal chips support them, and also disable ATSC in their Windows drivers. These limitations are of course not applicable to the Linux drivers. I don't know if there are cards out there that legitimately support only DVB and not ATSC. I do know that many support both.

If it helps you, my card uses a cx88 chipset and an nxt200x frontend. These seem to be pretty popular. The hardest thing for me to find was the firmware file dvb-fe-nxt2004.fw. You'll have to Google around for it. You also may have to set the card and tuner parameters if the card isn't well-known. Mine were card=34 and tuner=68 for an "ATI HDTV Wonder"-compatible card, but you'll need to do some research on your own for whatever card you choose.

I'm sorry for getting so technical here, but it took quite a bit of futzing to get my card to work, so, if I can help someone now or later, I want to take the opportunity to do so. I'll leave you with a hopefully helpful link: https://www.mythtv.org/wiki/ATSC

Linux support for digital video broadcasting

Posted May 30, 2015 23:59 UTC (Sat) by bradh (guest, #2274) [Link]

How much of this support is (re-)usable for "narrowcast" applications, including outside of embedded applications?

For example, a video source, perhaps audio, some metadata, all multiplexed together into some kind of transport stream over IP. I'm thinking presentations over a LAN, or perhaps a small UAV (although the same concepts get applied by the big kids: http://www.gwg.nga.mil/misb/).

Presumably this could be handled in userspace (e.g. VLC) if there is no hardware support, but as hardware support increases, use of that hardware might gain performance or power-efficiency.


Copyright © 2015, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds