I see that pretty much every vendor in GSM industry uses their own dialect in their core network products, I wonder why. What do they achieve from it while killing compatibility with other products from other companies?
Hi,
On Sun, Jul 17, 2011 at 08:52:37PM +0900, Seungju Kim wrote:
I see that pretty much every vendor in GSM industry uses their own dialect in their core network products, I wonder why. What do they achieve from it while killing compatibility with other products from other companies?
It's pretty simple: They ensure the operator is in a "vendor lock-in" and thus has to buy all equipment from the same vendor.
Thus, there is no real competition and the original "mix and match" idea as defined in the GSM specs is rendered void.
As an equipment supplier, you can then extort a lot of money from the operator, since switching from one supplier to another and replacing equipment is an expensive proposition.
What I'm personally surprised though is why nobody hires people (like us) to develop translators / gateways between those dialects. That's probably a hell lot cheaper than replacing your core network from one vendor with another vendor...
Regards,
On Jul 18, 2011, at 3:43 AM, Harald Welte laforge@gnumonks.org wrote:
Hi,
On Sun, Jul 17, 2011 at 08:52:37PM +0900, Seungju Kim wrote:
I see that pretty much every vendor in GSM industry uses their own dialect in their core network products, I wonder why. What do they achieve from it while killing compatibility with other products from other companies?
It's pretty simple: They ensure the operator is in a "vendor lock-in" and thus has to buy all equipment from the same vendor.
Thus, there is no real competition and the original "mix and match" idea as defined in the GSM specs is rendered void.
As an equipment supplier, you can then extort a lot of money from the operator, since switching from one supplier to another and replacing equipment is an expensive proposition.
What I'm personally surprised though is why nobody hires people (like us) to develop translators / gateways between those dialects. That's probably a hell lot cheaper than replacing your core network from one vendor with another vendor...
I think it is because of the user agreement, I do not think that mobile operators are allowed to trace the dialects. Have you heard of a game Starcraft? Its end user agreement says that users must not sniff packets nor make gateways that mimics their protocols. Of course source modifications are not allowed either. I think there are such agreement between mobile operators and equipment suppliers about which we do not know much. That also could be another reason why used network equipments are so hard to get.
Regards,
--
- Harald Welte laforge@gnumonks.org http://laforge.gnumonks.org/
============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)
On Mon, Jul 18, 2011 at 04:00:29AM +0900, Seungju Kim wrote:
What I'm personally surprised though is why nobody hires people (like us) to develop translators / gateways between those dialects. That's probably a hell lot cheaper than replacing your core network from one vendor with another vendor...
I think it is because of the user agreement, I do not think that mobile operators are allowed to trace the dialects.
I don't think you could do this, at least not under German law. Running a program and observing its input and output are always permitted, as long as you legally obtained the program.
How do you think that Samba would have ever become such a success, and unharmed by Microsoft? Because they couldn't do anything against people observing communications protocols.
Have you heard of a game Starcraft? Its end user agreement says that users must not sniff packets nor make gateways that mimics their protocols.
Not everything that people put in their EULAs is legally valid and enforcable.
Gentlemen,
What I'm personally surprised though is why nobody hires people (like us) to develop translators / gateways between those dialects. That's probably a hell lot cheaper than replacing your core network from one vendor with another vendor...
Hm, Consider this: For a network operator, the game is to get as close to 99.999....% uptime as economically possible.
Why? If something breaks, and you can't fix it within the hour, you'll lose your credibility, and your customers will eventually move to another network. So its really all about business continuity, reliability and risk management.
Hi Thomas,
good to hear from you!
On Mon, Jul 18, 2011 at 11:00:51PM +0200, Thomas Seiler wrote:
What I'm personally surprised though is why nobody hires people (like us) to develop translators / gateways between those dialects. That's probably a hell lot cheaper than replacing your core network from one vendor with another vendor...
Hm, Consider this: For a network operator, the game is to get as close to 99.999....% uptime as economically possible.
well, that (as well as the remaining argument) is true at least here in Europe. It's much different if you go to other places of the world...
This is the oh-so-familiar argument on why many companies use exclusively Microsoft, or everything Apple, etc...
ACK, it's the same story all over again. And still in each of the PC, Desktop, Server, Application Software, OS, etc. markets you see much more mix + match than in the GSM world... Even if you look at IP networks, not _everything_ is cisco, you commonly find heterogenous installations.
Regards, Harald
On Sun, Jul 17, 2011 at 8:43 PM, Harald Welte laforge@gnumonks.org wrote:
What I'm personally surprised though is why nobody hires people (like us) to develop translators / gateways between those dialects. That's probably a hell lot cheaper than replacing your core network from one vendor with another vendor...
Though greater interoperability contributes directly in reducing switching costs, manufacturers enforce their lock-in strategy using also license agreements.
For every degree of freedom adopters ask for (e.g., one more knob or switch, or vendor independence), they are willingly or not accepting one more quantum of liability. At the end of the day -- in a world where there's already very little accountability on the part of producers and products sold today -- adopters find themselves holding the liability bag. This can have a major impact on the operational risks.
There is an opportunity for us, if we deliver not only a gateway between network dialects but also confidence about the (updated) processes to which operators would like to entrust their businesses in order to overcome the vendor lock-in.
Regards,
--
- Harald Welte laforge@gnumonks.org http://laforge.gnumonks.org/
============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)
Ciao, alfonso
-- alfonso tweets @secYOUre blog http://Plaintext.crypto.lo.gy/
-
Alfonso De Gregorio -
Harald Welte -
Seungju Kim -
Thomas Seiler