* Posts by Christian Berger

4344 posts • joined 9 Mar 2007

BT pushes ahead with plans to switch off telephone network

Christian Berger
Silver badge

Re: Voice quality

There are some providers insisting on G.729, that's when you get shite voice quality.

Any semi-decent one will give you G.711 which is (except for a bit more latency) indistinguishable from ISDN. However even the latency should be much lower than 150 ms end to end. If it's not you or your ISP are doing something seriously wrong. Typical problems include not traffic shaping the Uplink and not prioriticing UDP.

Any decent telephony provider will tollerate no more than a single packet being dropped per 10 minute telephone call.

1
0
Christian Berger
Silver badge

Re: So what about the customers?

"If we're looking at VOIP and digital-only to the premises, who's going to pay to make my analogue interfaced cordless system a VOIP compatible one?"

First of all, my condolences for that piece of kit. You should sue the person who sold you that... but I digress...

In Germany the scenario for old, so called ANIS lines (essentially people who still rent their Dialphone for an Euro a month) is simple. You install a special line interface which is essentially an ATA so you can have all your analogue goodness like Impulse dialing, static, echoes and even semi-broken signaling so your answering machine will record some noise or beeps when the caller hangs up.

For people who want to have Internet along with ISDN, there's not much change. The most popular routers people buy in Germany already include a very decent VoIP stack you can plug your ISDN phones into (or even dialphones if you insist), and they even include a DECT base station. If you are one of those customers renting the CPE and you don't have the necessary equipment, you'll get CPE with at least one port for your dialphone. Some telcos in Germany, like Deutsche Telekom, are known for extremely shitty CPEs.

1
0
Christian Berger
Silver badge

Re: Digital Fibre Future

"Yes,all exchanges were digital since the mid 1980's,"

Sorry but that probably should be "all _new_ exchanges were digital since the mid 1980's".

In Europe development on digital exchanges started in the early 1970s when phone companies were hyped about computer powered switches. The only country I know of that saw significant use of those were the USA. The idea was that once you have such a system running, you could just replace the analogue switching matrix with something digital, and you get a completely digital system once that was more economical.

What they didn't take into account were the advances in microelectronics. While back in the early 1970s it was perfectly normal to have a computer with ferrite core memory, it was ridiculously outdated by the early 1980s when development was done. The result was that large parts of those switches were re-developed, based on microcomputers. Those switches then were completely digital and gradually came to service in the 2nd half of the 1980s.

Here's a commercial for a 1970s style analogue computerized switching system:

https://www.youtube.com/watch?v=QgB0KSjC2zg

https://www.youtube.com/watch?v=IbTGVN2VMnQ

Ohh and here's a BT film about their development of ISDN switches

https://www.youtube.com/watch?v=xy_6DL4haJA

There's even a 1984 Japaneese childrens programme about I(S)DN. Here's the German dub of it:

https://www.youtube.com/watch?v=4sCuN6TE8y4

2
0
Christian Berger
Silver badge

Re: Yeah right @Hoppy

"If I remember correctly, ISDN specified a 144Kb/s link, which could carry a 2 voice calls, each using 64Kb/s, and a 16Kb/s signaling channel."

That's correct, though those 2 64k channels could carry everything, even non-voice. It's signalled via the "bearer-capability".

"Also IIRC from my POTS training, analog phone lines used to have a filter at 8KHz, which was regarded as plenty high enough to carry voice communications."

Well there were early very long lines which used indutances on the lines which acted as a low pass filter, but extended the reach. I don't know exactly where that filter was.

However the actual limit was when carrier-wave systems were introduced shortly before WWII. Those stacked voice channels in frequency so a single coaxial cable could carry dozends of voice channels. So obviously you had steep filters to only give you a passband of 300-3400 Hz so they could stack more channels. (in fact there are reports about the stacking being changed during the day, so at night you actually got wider channels) Back in the days however you were likely to even get less if you had a worn out microphone capsule.

As for digital telephone networks they decided to use a sampling rate of 8000Hz as this allowed for affordable analog filters on both sides and was well withing the technical capabilities of the 1960s. The codec they used was G.711 which could be implemented fairly easily as it could be done by having some analogue circuity and an 8-Bit A/D converter.

ISDN actually had a special bearer capability for G.722 encoded audio which allows for frequencies of up to about 7.2 kHz to be transmitted over a single 64k channel. This caused quite some hype at radio stations, but ultimately fell into obscurity.

1
0
Christian Berger
Silver badge

Re: Mobile as the emergency option?

"Not really - you don't get location information from mobile."

That really depends on many factors. The interface towards the emergency services in Germany has ways to transmit either an address or a set of geometrical figures indicating the location. In the US, for example, virtually all phones have a way to capture a short burst of GPS data which then will be sent to the base stations in order to get a location fix for the phone which will then be transmitted.

1
0
Christian Berger
Silver badge

Re: Yeah right

"I know VOIP is better than it was some years ago, but it certainly seems that overall phone call quality has gone dramatically down over the years."

Well that depends on many factors. There are providers and PBXes insisting on the god awful G.729 for example, while any decent provider will use G.711 which is just as good as ISDN (but with a longer delay), good providers will support G.722 which does much better quality at the same bitrate. Another problem are really bad ATAs. Quality doesn't seem to correlate with price. The best ones (I've seen) for home uses are the "Fritz!Box" series from AVM, which you can get refurbished for about 70 Euros, but cost around 150-200 Euros for the top of the line model... which includes an internal ISDN port, a DECT base station, as well as a V/ADSL modem. Software support usually is several years for feature updates, and longer for security updates.

6
1
Christian Berger
Silver badge

It depends

First of all you already have that problem with regular ISDN, there the solution is simply to have a local battery backup... which your PBX will need anyhow.

If you still get a network connection depends on the way it's handled. For example classical ADSL tends to come directly from the old "switching office" where you have battery backup, so it should work fine. VDSL, particularly when done at the "curb" would need decentralized battery backups which may work. It won't work for vectoring as those boxes need _insane_ amount of power. If you have a dedicated fibre to your "switching office" to your home, it's likely to work. DOCSIS has many amplifiers and media converters, some of which are powered by the "groundstation" some are somewhere hidden inside your home.

The good thing about VoIP from a reliability aspect is that you just need any kind of decent Internet access. At work we've had many companies using even things like LTE when their wire based connection broke down. For a competent administrator it's easy to patch together a perfectly acceptable emergency solutions. This is far harder with ISDN as if your provider's ISDN switch goes down, you're toast and there's nothing you can do about it. ISDN equipment used to be highly reliably, however now 30-40 years into the lifetime of the equipment you find more and more failures, but no more spare parts.

So in short it's hard to say if VoIP will be more or less reliable given a certain situation. The main problem on current networks is that operators are trying out every new feature they can find. The result is that things like DTMF won't work, because one operator wants to do them as telephony-events, while the other one wants to do them inband (the saner alternative), and they somehow mess up signalling so both sides have different opinions on what's been negotiated.

4
0

OK, this time it's for real: The last available IPv4 address block has gone

Christian Berger
Silver badge

That's called 6to4 and already works nicely

It's for people who only have an IPv6 connection to connect to IPv4 hosts. AFAIK it uses some sort of NAT mechanism for this. It cannot be done directly as the legacy host would only get the truncated address and therefore couldn't reply.

Your suggestion would essentially be the same as IPv6, but with much shorter addresses. You'd still have all disadvantages of the switch, but without any of the advantages.

1
0
Christian Berger
Silver badge

Re: Time to claw some back

"What's the point in trying to claw back IPv4 addresses? It would not fix the problem, just delay it for another couple of years."

There's actually an interesting thought there. There are multiple groups of people who are "anti Internet".

One is the Facebook Crowd, they only want Facebook, not the Internet. Those people typically either don't use E-Mail at all or use one of the few largest mail providers.

The other one is the people being fed up with their ISPs meddling with the Internet and certain agencies sniffing it all, so they create their own overlay network using the Internet only as a transport network for their VPNs.

So there's a chance that in a few years people don't want the Internet any more.

3
48
Christian Berger
Silver badge

Re: IPv6 in the DMZ

Well actually many companies are already working on getting IPv6 in their internal networks, as those private blocks are already to small for them, but yes, if you are contempt with E-Mail and the Web, there is no need to have IPv6 on your internal network.

1
6
Christian Berger
Silver badge

Well first of all...

it can coexist with IPv6 and for most applications you can easily mix them. Essentially everything that runs though NAT also works through 6to4.

Also it's kinda hard to extend IPv4 to longer addresses in a compatible manner. Feel free to give some actual ideas. Putting everybody behind NAT is _not_ a solution.

10
2

Cisco snuffs Spark, renames it 'WebEx Teams'

Christian Berger
Silver badge

Multi-Standard RANs make a lot of sense

Particularly since since GSM, large parts of the base stations were implemented as an SDR anyway.

Besides LTE and GSM can perfectly well coexist on the same frequencies, if you manage to sync both basestations... which is trivial if they are combined. You can even ditch all that GSM fixed infrastructure and make the cells interwork with LTE more directly, greatly simplifying your architecture.

0
0

Surprise! Wireless brain implants are not secure, and can be hijacked to kill you or steal thoughts

Christian Berger
Silver badge

The question is actually rather irrelevant...

... we are not yet ready as we currently would still let companies do such implants. Once they become sophisticated enough to do complex things, you can bet that the manufacturers will use them for advertisement and other forms of attention monetarisation... just like they already did with Smartphones.

1
0

What Israel's crack majority-women Unit 8200 hackers can teach tech about diversity

Christian Berger
Silver badge

Re: Missed opportunity

This is a secret service, that's kinda the opposite of hackers. Those people use different kinds of force to concentrate information, Hackers share information.

Such a total conscription probably is even bad for the hacker culture in Israel as people don't get taught to think by themselves in military and secret services.

1
6

Intel's security light bulb moment: Chips to recruit GPUs to scan memory for software nasties

Christian Berger
Silver badge

Re: [C|G|F|S]PU Silicon shuffle

"Great, then we're going to call them Security Processing Units, add yet another $100 to the unit price and ship all that silicon anyways."

Seriously nobody would complain if they'd do it like that and sell that as an optional feature. It's shoving that "feature" down our throats that's the actual problem.

2
0
Christian Berger
Silver badge

Yet another itteration of the "anti-virus" concept

They'd gain more security if they'd remove their management engine and blocked the start of any office product.

12
1

Productivity knocks: I've got 99 Slacks, but my work's not done

Christian Berger
Silver badge

Re: Productivity

It's a bit like with messies who believe that if they just have enough things for organisation, they can do so easily.

Ohh and there are certain people who haven't learned to distrust marketing. So they actually believe that products like "Office packages" increase productivity, when they in fact just make a large part of your organisation waste their time on things they can't do well.

1
0
Christian Berger
Silver badge

Fascinating how a whole industry can be created...

...just by companies having incompetent enough IT departments to set up a private IRC server, or people being forced to use terribly bad e-mail clients which do not even quote propperly.

Seriously, if you want instant communications, there's XMPP or IRC, or the telephone. All three of which should be used in moderation, as their real time nature means that they are disturbing the recipient. (well maybe except for IRC which comes with a scroll buffer)

If you want something from a person, and it doesn't have to be _RIGHT_NOW_, E-Mail is your best chance to get it within a day or so.

1
0

Huawei promises to launch a 5G smartmobe in second half of 2019

Christian Berger
Silver badge

From what I've heard, the main innovation of "5G" will be...

... that it's essentially the same as LTE, but with every network internal message being now sent in a different format over web based protocolls.

Source, the Q&A session of this:

https://media.ccc.de/v/ARMP3D

1
0

Microsoft has designed an Arm Linux IoT cloud chip. Repeat, an Arm Linux IoT cloud chip

Christian Berger
Silver badge

Re: Embrace Extend Extinguish

Firefox now forces you to use PulseAudio.

2
1
Christian Berger
Silver badge

Re: DROP

Well either that, or it would be like with VBA or Active-X which everybody hoped they would drop it ASAP.

6
2
Christian Berger
Silver badge

Re: Remember when..

Yes, though the stronger vendor locks exist with SoC Companies like Broadcom or Allwinner, at least in the mobile market.

In a way it's like with home computers. We have lots of different vendors there, all with their own lock-in hardware, but most of them ran some version of Microsoft BASIC.

1
1
Christian Berger
Silver badge

Apparently the real news is that Microsoft freely licenses the chip designs

So the good that could come out of this is that manufacturers could use this as a single hardware platform. That's something that's desperately needed in the ARM world, as currently everybody needs to do their own Linux kernel if they want it to run on ARM.

1
8
Christian Berger
Silver badge

Embrace Extend Extinguish

It's a classical Microsoft strategy. They tried the same with the world wide web.

Though this is probably just a side project. The real damage is done by the SystemD/Freedesktop/PulseAudio people

18
6

'Well intentioned lawmakers could stifle IoT innovation', warns bug bounty pioneer

Christian Berger
Silver badge

My favourite would be the way it's done with electrical appliances...

... just have a set of evidence based measures the device has to do. For example there has to be input valiation which satisfies some constraints. Depending on your language, the compile could even check for such constraings automatically. So in the ideal case you would have a log from your compiler. If something bad happens you just have to show the log as well as the source code so you can proof that you did everything correctly, satisfying the rules of the time you shipped out the box.

For appliances this works a bit differently. For example there is a rule in the German regulations that no dangerous situation may occur if a single part fails. This is checked by laboratories looking at the schematics and trying to find the parts which would get closest to a dangerous situation. They then break those parts and test the safety again.

3
0
Christian Berger
Silver badge

"Financial pain is probably the only incentive. Or jail time."

That only works if you can still find that company or that company still exists then.

2
0

A developer always pays their technical debts – oh, every penny... but never a groat more

Christian Berger
Silver badge

Re: One way to avoid technical debt: experience

Well I don't think it's even possible to do decent software development in an outsouced environment. Software development without knowing in depth what the software has to do and what it's all about is utterly fruitless.

5
0
Christian Berger
Silver badge

One way to avoid technical debt: experience

Essentially most things developers do in the real world are utterly trivial. There rarely is an actual challenge. That's why developers often get bored and try to solve the trivial probems in more and more complicated ways. A good example are web applications. The web was made for (quasi) static "pages" of "rich text" with hyperlinks. Somehow people now try to cram interactive applications into this. That way, things that could be done in Delphi via the built-in "create a database application" wizzard within seconds, now take months.

If you have experienced programmers, which have worked in different areas, you can reduce your technical debt. Just let them choose the propper tools for the job. Let them build prototypes so the users can already play with the product long before any details are set in stone.

9
3

Imagine you're having a CT scan and malware alters the radiation levels – it's doable

Christian Berger
Silver badge

Can that even happen?

I mean an X-Ray CT needs a certain range of radiation power, after all the stronger your source is the more expensive it will be. Shouldn't there be hardware saveguards in place to keep the (moderately expensive) X-Ray tube from being overloaded? I mean every CRT TV has internal overvoltage and overcurrent protection to make sure the TV turns off immediately in case the CRT is operated out of spec.

It seems to me that this could just be some cheap alarmism to attract attention. Just like we had with the guy who claimed that running your own firmware on a printer would cause fires.

Obviously though you should never run any part of that equipment, not even the GUI, on unhardened Windows.

0
0

Skype for Business has nasty habit of closing down… for business

Christian Berger
Silver badge

Re: Suggestions?

There's the telephone network which slowly even becomes video capable.

0
0

No password? No worries! Two new standards aim to make logins an API experience

Christian Berger
Silver badge

The proponents of biometry would say...

... that somehow biometry has magic sensors which cannot be fooled and those sensors somehow securely talk to the authenticator and authenticate themselves, etc.

How does this work? Magic!

3
0
Christian Berger
Silver badge

Well it's probably about that data protection stuff in the EU

Essentially the EU wants to make it harder to track you. Unfortunately there's a loophole. If you have a user account they can still track you if you consented during login.

4
2
Christian Berger
Silver badge

WTF!!!! There are already several standards

HTTP has well working ways to authenticate a user, so does TLS. In fact TLS client certificates could even be simpler and more than secure than anything else, if only browser vendors would make them usable.

3
1

Gemini: Vulture gives PDA some Linux lovin'

Christian Berger
Silver badge

"Planet is a tiny company with a shoestring budget."

Well yes, of course... but so is "Open Pandora" who are working on the Pyra and already released the Pandora, both made with open SoCs so a propper Linux was available from the start. (way before Android)

What Planet is doing is shipping their devices before they are ready. I can understand why someone would do that, but it's a strategy that will lead to some unsatisfied customers. After all it can be more frustrating to have a device you cannot use and eventually will have to update, than not having a device at all.

To their credit they have done some things to even make Android usable, namely offering a rooted version of it. If they would now release the apk of their new keyboard app, the device would be good enough to critique.

3
0
Christian Berger
Silver badge

Re: Think it could be simpler

"more choice of phone etc."

Yes, but virtually no phone out there comes with any kind of normal operating system. Even the x86 phones are not PC-compabible so you cannot install $Operatingsystem of your choice.

Android is not what the people buying it, bought it for. It's a stopgap solution until the actual Linux is there... or until the Pyra ships.

6
0
Christian Berger
Silver badge

Re: Nokia communicator

Well the Nokia communicator was the ideal form factor, as it kinds removes the problem of needing a phone interface by giving you essentially a dedicated phone.

1
0
Christian Berger
Silver badge

Good question

"can you make/receive calls/SMS and use mobile data if you're running Linux?"

That's a good question, I haven't gotten the provided Linux image to do anything usefull. I'd say "probably yes", but maybe not comforably for the first couple of versions. Again firmware wise this is far from complete.

2
0
Christian Berger
Silver badge

Re: Linux: all the tools are Windows-based.

" just in case one Penguin's last copy of Windows was pre-NTFS."

Even with NTFS the default Windows ZIP archiver cannot deal with files larger than 4 Gigabytes. So I had to unpack it on Linux, shift it over via an NTFS disk and then copy it to the system disk.

2
1
Christian Berger
Silver badge

Re: It's still rather disappointing

Well I ordered German. However the keyboard support currently is less than stellar. Still miles above any kind of screen keyboard.

1
0
Christian Berger
Silver badge

It's still rather disappointing

I mean it will probably improve gradually. So far the keyboard only works with the English layout, which if you have the German version, means that many keys are labelled "wrong" and you need to guess punctuation (which is accessed via a special modifier key). Yes there is an updated keyboard app, but that's only available via a third party "Appstore" which I'd rather not use. Just offering the apk as a download would greatly improve both security and usability of the device.

On the plus side, the manufacturer offers a rooted image, so potentially it could be possible to get iptables on it so you can filter your outgoing traffic to prevent it from talking to untrusted 3rd parties. So far it works nicely as a mosh client.

5
0
Christian Berger
Silver badge

Re: Linux: all the tools are Windows-based.

Well they are essentially tools the chipset manufacturer provides. They do work, though bizarrely not all Windows software can deal with files larger than 4 Gigabytes.

6
0

Want to terrify a city with an emergency broadcast? All you need is a laptop and $30

Christian Berger
Silver badge

Re: "... warning sirens...only truly reliable method..."

Well but natural disasters usually don't jam radio signals. (at least not continuously) However things like earthquakes can easily break cables.

6
0
Christian Berger
Silver badge

That wasn't a design goal

Seriously the design goal is that in case of an emergency there will be an alert. False alerts are not really a big problem, unless they actually happen rather often. So for example using TLS as part of your protocol, would be a problem as there is a chance it might fail because of expiring certificates or because there was some intermittent power outage causing the clock to be wrong.

10
3

Russian regulator asks courts to disconnect Telegram

Christian Berger
Silver badge

Re: Peer-to-peer voice text encryption

@Mike16: Well you are mixing up a lot of things.

First of all if your landline provider is using codecs like G.729 you should seriously be considering to swap them for someone who knows what they are doing. There is no reason to use that codec as the licensing costs are far higher than the bandwidth costs. Any sane telephony provider will give you G.711 (either a or µ depening on the continent) which is the same as used on ISDN.

Then there's really bad CPEs. One of the main problems with VoIP is that both the transmitter and the receiver need to run at precisely the same clock. That either requires you to have a precise crystal oscillator, or to estimate and compensate your clock error via NTP. For some reason many CPEs do neither of those. So you'll end up with your transmitter transmitting frames with 8001 Hz sampling rate, and your receiver playing them with 7999 Hz. After a short while the timing difference will have made up a frame, and a frame gets dropped... many modem standards don't like that at all.

So modem transmissions do work, if you have a decent CPE and a decent voice provider. In fact on many voice providers you can even use ISDN transparent data transfers. Most protocols based on that can easily cope with the frame slips mentioned above, so that's even rather solid with cheap equipment.

However I'm talking about something else here: Imagine you have a mobile phone to mobile phone phone call. Both phones speak, lets say AMR as a codec. In the past this would have been transcoded to G.711, sent to the other carrier, and transcoded back to AMR. That is however expensive (proprietary voice codecs cost a _lot_ of money per channel) and decreases the quality of the call. Therefore phone companies try to avoid this more and more. Therefore they try to just send the data through verbatim.

Usually your codec turns voice into bits. Who says you need to actually encode voice? For the network bits are just bits. So if you bypass your voice codec and just send raw data, you will get those data on the other end. (provided there is no transcoding)

So essentially you'd start your call, and for the first second or so you transmit some bit pattern which would decode to some non-annoying noise. You can do that on both ends and detect a compatible peer. Then you know you have a bit transparent channel you can negotiate your encryption on. Once you are finished, you use a codec with a slightly lower bitrate and use the rest of the bits to work on renegotiating the next key while you encrypt your voice data.

The best thing about this is that your call will just look like any normal call. Your telephony provider has no idea its encrypted as the signaling is normal. This also would automatically work without any manual negotiation. If you happen to dial a compatible phone, it'll all happen automatically.

1
0
Christian Berger
Silver badge

Re: Peer-to-peer SMS text encryption

Well you can't send photos via SMS, only via MMS... which is probably the second most expensive way to send any kind of data.

What would be interresting, in theory, would be to send it as a voice call. Those are more and more likely to be bit transparent as inter carrier links get converted to VoIP which makes it easy to support all those wierd codecs like AMR.

3
0
Christian Berger
Silver badge

Re: Peer-to-peer SMS text encryption

Well the problem is that SMSes are fairly small so you won't be able to get propper encryption. However you could have a pre-shared key and use some symmetric encryption methode like AES for it.

You simply cannot send a 1024 bit key when you only have 1120 bits for your whole message.

4
0

IBM swings shrink ray from workforce to mainframes

Christian Berger
Silver badge

I'd like to have an honest non-marketing answer to the question...

... what's so special about those boxes?

I mean you can run Docker images on cheaper hardware, too.

5
0

Company insiders behind 1 in 4 data breaches – study

Christian Berger
Silver badge

Well of course...

if there is nobody inside the company who collects data, nobody outside can steal it.

It's not the leaks that are the problem, it's the collecting. If you business model is based on collecting data you normally shouldn't have, maybe your business model needs to be outlawed.

2
4

There's security – then there's barbed wire-laced pains in the arse

Christian Berger
Silver badge

Unfortunately

Many IT-departments choose to have neither of those:

For example ours forces us to use insecure systems (we have to use Acrobat Reader for PDF, as well as Office Products) it filters outgoing E-Mail for document types like .wav. It's probably spending a lot of money for "security solutions" which do nothing, and their e-mail solution can't handle mailboxes larger than 2 Gigabytes.

The optimal solution changes depending on what department you are talking about. For an office department you might be able to just lock down Windows installations, but for technical departments the easier and much more secure way is to use Linux or some BSD. Nobody in a technical department will care about compatibility bugs in evince or even consider sending HTML E-Mail.

0
0

Modern life is rubbish – so why not take a trip down memory lane with Windows File Manager?

Christian Berger
Silver badge

There used to be a time...

...when GUI designers actually cared about usability.

28
0

Forums

Biting the hand that feeds IT © 1998–2018