One of the approaches compartimentalization of embedded systems is
MILS (wikipedia),
it's e.g. followed by EURO-MILS and D-MILS.
A use case in cars is given here (in German).
Disclosure: I'm with EURO-MILS.
Some things that might not immediately turn up in google for "open source c++ numerical library" but maybe also of interest: symbolic computing GiNaC, or, into another direction, computational fluid dynamics finite element tools like OpenFOAM.
Actually in Verisoft XT the successor of Verisoft now real C code is verified, including e.g. PikeOS with PowerPC assembly in a system developed to DO-178B airplane verification. In that project an important tool for C code verification is VCC developed by Microsoft Research, source included at that site.
The Patent Office has today announced a series of workshops to examine the possibility of a more concrete definition of "technical contribution" in the Software Patent directive (NB: Registration deadline 18 February)
More see FFII UK.
There is wide consensus even in the patent community that patents work most successfully in chemical and pharmaceutical industry, less well in mechanical industries (Kingston 1997, "Patent protection for modern technologies", Mansfield 1986 "Imitation costs and patents".).
Recent empiricial research suggests patents are even less successful in software. This is because software is different. Or search for "while statement" here.
To illustrate cumulative innovation complexity-wise, the 2.6.8 kernel has 300,000+ IF statements, a BMW sedan car has a complexity of "only" 15,000-18,000 pieces (Mr Blabst BMW press department), a typical drug consists of 10-100 atoms.
In other words, the position that there are fields of human activity that the patent system is not so well tailored for is well-defensible even without killing the entire patent system.
Sidenote: there have been a lot of quirks and surprises in the past with the Software Patent Directive, so get prepared to the thought that next week's Brussels/Berlin demos
are still needed and useful politically (only one scenario of total fiasco has hopefully been avoided, there is not yet a renewed referral in Parliament nor B-iten in council).
Japan, Australia already have software patents. AFAIK, India,
New Zealand, most of Mercosur do not have
them, though some laws are under discussion too (IN, NZ).
Here you find a
(somewhat Euro-biased) list of FFII regional groups, join and take the initiative (not all mailing lists are very active)
Thanks.
an option is to go via US Congress to disallow patents on computer programs (as hopefully Europe, India can do). Join eg https://lists.ffii.org/mailman/listinfo/us-parl for that. (Another helpful thing would be patent situation analyses, eg on database patenting).
https://lists.ffii.org/mailman/listinfo/us-parl/ offers a mailing list for US FFII supporters. Honestly it is not yet very populated or active, but it's an open door and if you want to become active and have some medium-term stamina then that's a start. Similar lists exist for European country-codes at-parl,be-parl,cz-parl,... or e.g. in-parl if you are e.g. from India.
There are about 30-40% applications inside (easy to spot, everywhere where it says 'Anmeldung'). However, with a granting rate of over 50% it seems reasonable to take some of these into the balance when evaluating future risks.
Also note that the US did not have software patents during the 70s/80s when many of today's big players grew up. Nor does India with its dynamic industry have software patents.
As you can see on a newswiki in addition to the abstentions
of Belgium (5 votes), Luxembourg (2 votes), there are some positive statements
from some eg Italian, Slovenian, Spanish etc politicians too, but it is very much in your interest keep the in
touch with your government today and on Monday too (Discussion/Voting in the Council scheduled for
Tue 18 May; calculate for some time for transmission of your local govt's opinion to Brussels representative!). More help (including pointer to irc) here.
In due fairness, spreading had not been disencouraged and it is just the official release at http://swpat.ffii.org/news/04/amaz0125/index.en.ht ml that had just been a bit slow:-)
Extracting the address from an email, web page etc is not in the first claim which stands alone and has been granted by EPO. In claims 2-6 where "parsing an email" etc is mentioned, there is no mention how it is done.
But FFII's opposition is not based on non-novelty or lack of inventive step (except that Fleurop's opposition is fully endorsed) but rather on the ground that software as such is not patentable under EPC.
Slashdot Mirror
One of the approaches compartimentalization of embedded systems is MILS (wikipedia), it's e.g. followed by EURO-MILS and D-MILS. A use case in cars is given here (in German). Disclosure: I'm with EURO-MILS.
If you are in the EU, there is currently a public EU consultation on open access (deadline 09 September). This was also mentioned on German heise.de earlier this week.
On the regulatory side, for networks the NERC Reliability Standards for the Bulk Electric Systems of North America address similar concerns (including cyber security) in electrical grids. For highly integrated systems MILS kernels are an engineering solution e.g. to keep actuators and monitoring subsystems apart.
Some things that might not immediately turn up in google for "open source c++ numerical library" but maybe also of interest: symbolic computing GiNaC, or, into another direction, computational fluid dynamics finite element tools like OpenFOAM.
(Or more generally see the main Verisoft page. When posting on C verification perhaps one also should mention Caduceus)
Actually in Verisoft XT the successor of Verisoft now real C code is verified, including e.g. PikeOS with PowerPC assembly in a system developed to DO-178B airplane verification. In that project an important tool for C code verification is VCC developed by Microsoft Research, source included at that site.
There's a [mailing list http://lists.ffii.org/mailman/listinfo/in-parl/] here. The Indian ordinance needs to be ratified by Parliament yet.
The Patent Office has today announced a series of workshops to examine the possibility of a more concrete definition of "technical contribution" in the Software Patent directive (NB: Registration deadline 18 February) More see FFII UK.
Recent empiricial research suggests patents are even less successful in software. This is because software is different. Or search for "while statement" here. To illustrate cumulative innovation complexity-wise, the 2.6.8 kernel has 300,000+ IF statements, a BMW sedan car has a complexity of "only" 15,000-18,000 pieces (Mr Blabst BMW press department), a typical drug consists of 10-100 atoms.
In other words, the position that there are fields of human activity that the patent system is not so well tailored for is well-defensible even without killing the entire patent system.
Sidenote: there have been a lot of quirks and surprises in the past with the Software Patent Directive, so get prepared to the thought that next week's Brussels/Berlin demos are still needed and useful politically (only one scenario of total fiasco has hopefully been avoided, there is not yet a renewed referral in Parliament nor B-iten in council).
Japan, Australia already have software patents. AFAIK, India, New Zealand, most of Mercosur do not have them, though some laws are under discussion too (IN, NZ). Here you find a (somewhat Euro-biased) list of FFII regional groups, join and take the initiative (not all mailing lists are very active) Thanks.
Internationally.
an option is to go via US Congress to disallow patents on computer programs (as hopefully Europe, India can do).
Join eg https://lists.ffii.org/mailman/listinfo/us-parl for that.
(Another helpful thing would be patent situation analyses,
eg on database patenting).
Hmm, I guess (correct me if that is wrong) that plenty of MEP assistants are in Brussels the following week anyway.
Do you have time to organize anything at FOSDEM? Would be very welcome. Please contact:
holgerlists at blasum.net (who will then try to route to another person who had a simliar idea), subject FOSDEM.
Thanks.
https://lists.ffii.org/mailman/listinfo/us-parl/
offers a mailing list for US FFII supporters.
Honestly it is not yet very populated or
active, but it's an open door and if you want
to become active and have some medium-term
stamina then that's a start.
Similar lists exist for European country-codes
at-parl,be-parl,cz-parl,... or e.g. in-parl if you
are e.g. from India.
There are about 30-40% applications inside (easy to spot, everywhere where it says 'Anmeldung'). However, with a granting rate of over 50% it seems reasonable to take some of these into the balance when evaluating future risks.
If elsewhere in Europe (where legislation is hot), here are some mailing lists.
E.g. Robert Hunt, Philadelphia Federal Reserve Bank, sees the the market equilibrium to be bad for software when patents come into play (he also quotes model calculations).
Also note that the US did not have software patents during the 70s/80s when many of today's big players grew up. Nor does India with its dynamic industry have software patents.
Software patents party trends, candidate statements and voting behavior for many European countries (it's a wiki, add your country if you feel it's missing ;)
+49-174-7313590 (gsm,myself)
Irc.
(apologies for the previous accidental ac dupe)
As you can see on a newswiki in addition to the abstentions of Belgium (5 votes), Luxembourg (2 votes), there are some positive statements from some eg Italian, Slovenian, Spanish etc politicians too, but it is very much in your interest keep the in touch with your government today and on Monday too (Discussion/Voting in the Council scheduled for Tue 18 May; calculate for some time for transmission of your local govt's opinion to Brussels representative!). More help (including pointer to irc) here.
In due fairness, spreading had not been disencouraged and it is just the official release at http://swpat.ffii.org/news/04/amaz0125/index.en.ht ml :-)
that had just been a bit slow
news,
see also some background.
Here is the filing.
But FFII's opposition is not based on non-novelty or lack of inventive step (except that Fleurop's opposition is fully endorsed) but rather on the ground that software as such is not patentable under EPC.
Bu yes, there are..