These are my own personal views and not those of my company Phaedrus Systems. www.phaedsys.org which is where the full version of this column resides under the Technical Papers button.

 

 Last month’s ESS went well and was worth visiting as a couple of thousand of you know.  A focal point for UK industry and the largest conference of it’s type in the UK. Those of you who don’t go don’t know how much you are missing.  Make the effort to go next year.

 

One event that will be of interest to many is the MISRA Forum in Coventry on the 29th November, C, C++, Autocode and safety panels will be reporting. The C++ team might actually have a document close to a release.   The C group are of course starting MISRA-C3 so now is the time to make suggestions either at the event or on the forum.  See www.misra-c.com  there are several tracks and a trade exhibition.

Recently I mentioned Patents, an important topic for new technologies, and that a US survey discovered that more money is spent litigating them than is gained from them. This is due to the US granting patents first and arguing later. The proof of this is at http://www.patentlysilly.com/  this you would think is a spoof comedy site until you realise it has links directly to the real patents at the US patent office.

 

However if you have a new “something” it is always a good idea to check if there are existing patents. And James from http://www.FreePatentsOnline.com contacted me to advertise this free patent search web site after the last column. I mention it as this sort of web site would be a good first step for those of you without a patent or legal department. I know of many small UK companies working on new ideas.  

Unfortunately the lawyers seem to intrude far more these days. There is an on going bun fight in the open source area about the new GPL (Version3) UK-Linux watch has story with lots of links and a pdf explaining all the differences between GPL V2 and V3 http://www.linux-watch.com/news/NS8508141327.html?kc=EWKNLLIN100207STR4   Also, according to eWeek a survey by Evans Data released Sept. 25, suggests that  less than 10 percent  Open Source Developers have moved to GPL3 and that whilst 60% say they won’t move this year over 40% have said they will never move  to it!  Apparently the major area of disquiet that GPL3 has a clause that forbids licensees from bringing patent infringement suits, directly attacking the alliance between Novell and Microsoft.   It is going make sorting out Open Source licensing more time consuming and expensive than buying normal commercial software!  At this rate the Great Satan (Microsoft) only has to sit back and let them all shoot them selves in the foot.  Perhaps the Open Source party is now coming to a close or possibly just a change to the next level.  The internet survived when the commercial world took over from the geeks as the prime users despite all the doom merchants. 

 

I think things evolve, often out of all recognition and Open Source is one of them that will survive but not as originally envisaged. Ask the US security people in the 70’s if they thought people would do their shopping over DARPA net.

 

Regarding changes and the future MISRA-C will be looking at C99 rather than C90 as most C compilers in the embedded sector are now C99 minus rather than C95 plus. Mainly because most compilers are now starting with a C99 parser.   Whilst hunting round for some information I came across the link to  http://www.research.att.com/~bs/bs_faq.html  Which is Bjarne Stroustrup’s fascinating personal insight to C++ and related issues.   It is current (as of November 2007) and looks to be regularly updated.  This page is essential for anyone working with (or near?) C++ it also kills many myths. 

 

I was sent recently the email below as both the recipient and I were some what stunned by the content and the implications. I have changed the names to protect the recipient NOT the Sender.  No, the * are not a 1 for 1 letter replacement! So no trying to guess the names

Here we come with status of the Target build issues
Real compiler errors, I believe due to *********** C parser being stricter than ********* on ANSI C compliance (hundreds in the build log supplied by *************)

Example:
Error ABC123, expression must be a modifiable lvalue (200 cases)

We will come back to this, if **** can supply less strict parser?
Is it possible to suppress these errors?

In other words: “Can you get us a less strict error reporter so we can ignore the errors?”  Fortunately this is not of critical software just some consumer stuff so it will only cause hassle to users not injury or danger.

 

The discussion I had with the person who sent it to me was “what would we have done if their project had been on a critical system?”  As professionals we could not have turned a blind eye.

 

What can you do, as a professional, when you find safety is being ignored. It’s great fun watching the good guys win eventually when it is a film but the real world does not work like that.  They don’t do films where the whistleblower looses job, house, family and future and The Company carry on making record profits.  Neither do they show what happens the day after the title credits role, and the hero has to come back to real life with a mortgage payment due and potential employers asking why did you leave your last job?

 

One item on standards where I can name and shame because it is on at least two public web sites for at least three weeks… I thought it was a spoof when it was sent to me. It is a job advert from an agency.  This is more standards of basic English and communication. Quite apart from care in what your company sends out.

 

It is a job advert from the agency  Corporate Resourcing Solutions Ltd. It  was apparently posted on their web site and the Electronics Weekly web site see: http://www.electronicsweekly.com/Jobs/2007/10/10/114161624/european+sales+manager++distributiors
+sales+bdm.htm

It is for a European Sales Manager at 25 to 30K in Hampshire.  Read the Job description (exactly as it appears on the web site)

This company is responsible for the design and manufacture of instrumentation and commercial electronics systems/products European Sales Manager

In this role you will be responsible for the developenmt of sales throgtuh disisye int eh UK and Europe.the roel will reurie some tevek to tehs elcoaitons. You will also be repsos mfor the devleiolenj fgof a starteged for the groer of proedyd sdlwes into tehse regiosn. You will aolso be rewuiire to rpeoei Techilc support an dtarainf to bioth dierctri cusosmer as well as distribtuiirs.

HND/HNC Electroncis background,
Corporate Resourcing Solutions Ltd is acting as an Employment Agency with regard to this vacancy

Before you ask we did check to see if it was in Klingon and I stored a copy of the web page just in case it gets denied later. I can understand, as a dyslexic, it is easy to miss spell words or have some mistakes even with the spell chequer but the job description above does not even have the excuse they had the US rather than the UK spell checker: a common mistake.  This shows a complete lack of care or any sort of standards.   I know “kids these days” use txt spk but I don’t think it even qualifies as that.

 

If you can actually translate it in to intelligible English do let me know what the job description really said. If you are interested do contact Peter Starling at CRS with reference PS613EW. On the other hand I should like to hear from Peter as to how that advert actually got published in that state.  Though I don’t think I shall be using CRS to find staff for me it is not the company image I want to project.

 

Over the last few months I have mentioned WiFi and the possible dangers of permanently living in an electromagnet soup both at home and at work.  Well there are lots more reports both for and against mobile phones (and masts) and wifi etc This looks remarkably reminiscing of the smoking/cancer scenario over the last five decades.  Most of the reports one way or the other all seem to have support from the side advocating the stance the report has. They would say that wouldn't’t they? So we are no wiser and by the time we do get concrete results 80% of the population will have been the long term guinea pigs.  So I turn off my WiFi except when I need it, I don’t use a mobile phone except when I want to send or receive calls and I object to any mobile mast within a kilometre of my home but get annoyed when I don’t have good mobile reception. I suspect that when the do know for certain shall be past caring.

 

It seems that the majority of the public are past caring. The Guardian In September  http://blogs.guardian.co.uk/science/2007/09/every_summer_a_university_camp.html  carried an item about University science fairs and public apathy.  One comment was that many of the movers and shakers in today’s technology were inspired by a trip to a museum, an event or being given something technical to make or use.  It seems today the kids by and large are not interested in doing so much as being passive users.  OK so we do have or rather had a lot of interest in games programming but little else in the science or technology fields. Robot Wars did stimulate interest in robotics all be it arguably for the wrong reasons.  However will the other TV shows like the Gadget Show are for users of the technology not how it and the technology work.    In a world that runs on technology why can’t we get youngsters interested in developing it rather than just using it?   Something to think about over the Christmas break.

 

Another one to think about that amused me. Apparently, according to the Telegraph http://www.telegraph.co.uk/money/main.jhtml?xml=/money/2007/09/12/cndsei212.xml
  A team of researchers and engineers at a UK division of Franco-German aerospace giant EADS has developed what it believes is the world's first hacker-proof encryption technology for the internet…….   OK when I see the words “hacker-proof”, “encryption” and “internet” in the same sentence I buy the following days edition to see who cracked it. Will they never learn?  Nothing is hacker proof.  If only for the simple reason that whilst statistically there may be 25 billion combinations they might strike lucky in the first five attempts.  I recall a security demo where some one was going to hack a server live at an event.  It took the hacker about 1 minute. How?  The machine was secure the “hacker” just got drinking with some of the support team the hotel the night before without them realising exactly who he was …. 

 

If you are into software security then you might be interested in this US Report on Software Security Assurance  http://iac.dtic.mil/iatac/download/security.pdf   It is long at about 400 pages and full of acronyms but it is comprehensive.  Of particular interest is the section on designing secure software and the table summarising what the requirements are for secure software….  As with most guides for safety (and less so for security) the rules are common sense and would be sensible to include in most software developments.    The reason why I say “less so” for security is that some times security is driven by irrational fear and paranoia rather than solid Engineering

 

You all know I bang on about certification for software Engineers but it seems even the US is pushing it.  According to the IEEE computer society “There is a growing need for professionals in the software development field to be certified. Decision-makers at many organizations look at a software developer's credentials before making selections for key positions and projects. Certifications signify that a person has demonstrated their capabilities to perform well in their profession.”    In the UK the IEE and BCS do various courses but the C.Eng is the one to go for it the UK.   More and more I am seeing a push towards certification so if you get a quiet moment over Christmas hit the IET or BCS website and have a look at the forms. It’s all tax deductible. It doesent matter which you go for. Ie expect the BCS and IEt to mege within the next decade anyway.

 

One final thing to muse over. . Is it impossible to build a warranted product from components that disclaim any sort of guarantee? http://www.embedded.com/columns/embeddedpulse/202100611?_requestid=833522

An interesting thought, the author of the article, Jack Gansle, has been in embedded systems for many years and knows what he is on about. The article discusses a report on Software for Dependable Systems that states “Clearly, no software should be considered dependable if it is supplied with a disclaimer that withholds the manufacturer's commitment to provide a warranty or other remedies for software that fails to meet its dependability claims."   Now so far I have not (ever) seen any software with out a disclaimer that it is even fit for purpose let alone dependable.  You may want to think about the implications if you have to start giving some sort of assurance your software will do what you say….  Strangely most consumer items containing software (phones, MP3 players washing machines) you can take back as they all have warranties and anyway consumers have legal protection. So why not other software or embedded systems that are not “just knocked out” for the consumer market? Surely there is more care taken over these items?  This is where liability starts to creep in again.   This links back to being professional.  Some one is going to have to sign off the software….. For the last decade or so I have been saying that, in general, the requirement for Chartered Engineers in software will be forced in by the insurance industry not from within. 

 

With that happy thought I wish you all a happy festive season no matter which festival you are celebrating, if any.  The New Year looks to be very interesting for the industry in general.

As usual any praise, death threats or other comments to chills@phaedsys.org. All messages are read even if I don’t manage to reply.