Instead of going to the IIW yesterday, I got pulled into a meeting to help work through some client issues.  The actual problems were completely tangential to my role on the project, but given my background I was happy to help. 

That morning meeting became contentious.  After the fifth time of hearing the same inadequate solution with a dose of attitude . . . well . . . I don’t cotton to that.

Kind of a mess, really.  I wouldn’t have done it at all except that I was helping out a friend.

And now, off to the IIW 2009A, day 2.

It runs Monday, May 18th through Wednesday at the Computer History Museum in Mountain View.

I’m a bit overbooked this week.  The song remains the same.  I’ll be there as much as I can and at the dinner Monday night at the Tied House.

If you want to meet up, contact me at robert at rwandering dot net or use the form at =rwandering.

I’ll post more on this later . . . as time permits.

I like those guys.  They are smart and do excellent work.

Anyway, they released the sample earlier today.  From their site:

As of version 1.5 (which is currently in beta), the world’s coolest spreadsheet can use Digipede Network grid computing to distribute and execute workbooks in parallel. The example on the Exchange is based on the excellent IronPython sample created by Robert W. Anderson of Digipede. The Digipede Network is a brilliant way to get distributed, parallel computation on Windows. It only took a few minor changes to convert Resolver One to run on the Digipede Network and to get the IronPython sample to execute Resolver One workbooks.

Giles gives some more background to the path that got us here on his recent post, Resolver One and Digipede.

The combination of our two products offers a pretty elegant solution.  Like I said before,

Try doing that with a spreadsheet or grid that isn’t based on .NET . . .

. . . like Excel and Windows HPC Server.  No, don’t. Trust me.  It is really hard, complex, and brittle.

I had hoped the transition to rwandering.net would be mostly automatic, but it seems that many feed readers ignore the “permanent” in the 301 redirect. That is, they are happy to redirect but ignore the fact that the feed URL changed. So when the old feed stops redirecting, b’bye.

The feed for this blog is http://feeds.rwandering.net/rwanderingMain.

I am making this statement for completeness.

They used my IronPython Worker sample and customized the front-end Python code, leaving the C# adapter as-is.  With very little coding they had an elegant grid-enabled spreadsheet.  Try doing that with a spreadsheet or grid that isn’t based on .NET . . .

Giles said they will have support for this in Resolver One 1.5, coming out in the next couple of weeks.

Very cool. 

I’ve just installed Resolver One to take a closer look.  Already I’m impressed, but I’ll leave that for a future post.

Your installers — and core components for that matter — shouldn’t be performing checks that preclude new versions of Windows.

This is particularly relevant with a new version of Windows coming soon.  In fact, I’m writing this because recently someone told me that his team’s software wouldn’t install on Windows 7.  The release engineer fixed it by increasing the high version limit.

That is the wrong approach.

The right approach is to eliminate any upper bound on a version limit.  The premise is that developers shouldn’t assume that their software won’t work on future versions of Windows. 

Windows Logo requirements have required this for some time, and in fact there is a certification test called the HighVersionLie that tests for just this case.  The test sets the Windows version to an artificially high number to see if the tested software still installs and runs.

BTW: One reader from Microsoft told me a similar story at last year’s PDC – this posting was prompted by a different conversation altogether.

Unfortunately, Chrome doesn’t support WSR.  According to Rob Chambers this would be easy for Google to do, and I suspect it is just an oversight on their part (both in terms of making their software more accessible as well as following Windows best practices).

Google:  when are you going to put the effort into this?  The Chrome 2.0 Beta doesn’t do it either.

Rob Chambers: how easy is this really?  You also said that Firefox does support WSR – maybe it does, but not in Google Docs.

So now, I’m using IE8.  Google Docs with WSR works great there.

This should be transparent, but if I lose you in the process, you’ll know why.

I have been playing with IronPython a little.  With the release of Digipede Network 2.2, I am now able to post the sample I wrote.  It shows how to distribute IronPython objects on the Digipede Network.  You can find it on the Digipede community site.  See the posting there for details and download instructions.

The sample uses IronPython 2.0.1 and the included version of the Microsoft Dynamic Language Runtime (DLR).  While I focused on IronPython in this sample, it would be pretty easy to expand it to support other DLR-based languages.

Comments welcome.  I am specifically interested in feedback on DLR integration and initializing ScriptScope objects for each worker thread.  It seems that I should be able to do some of this only once at global scope.

By the way, one thing I like about this sample is that it shows how to keep user code completely de-coupled from the Digipede Network while still taking advantage of our deployment and payload distribution model.  This has always been supported by the Digipede Network, and this makes a good example.

I have been using Dragon NaturallySpeaking for several years now.  Every new edition has gotten a little bit better.  And I’ve found it quite useful except that the last two versions have tended to get corrupted databases.  And now, I will very commonly get some error that causes it to lock up.  I have found that it tends to fail if I switch around between different applications.

Of course, I am an avid multitasker, so these errors has become really irritating, and essentially stopped me from using voice recognition.  I just can’t commit to staying in a single application for very long.

Today, I started using the speech recognition built into Vista.  While I am struggling a little bit with differences in the verbal commands,I am already sold.  Why?

• It feels much faster.
• I can switch between applications with no problems.
• The recognition engine is doing a terrific job with very little training.

I know that while I continue to use this I may come up with some problems, but so far I’m blown away.

Plus, Vista doesn’t try to sell me things through its auto-update feature.

I don’t ever want Live Messenger to interrupt my work with that dialog.

If connectivity fails, use a notification balloon, or better yet, do nothing.  The red X on the tray icon already signifies trouble.  Let me drill down to find out more info.  This is a bad design decision.

If I dismiss the dialog, it will happen again.  And if I don’t dismiss it?  Well,

I left my machine from 9pm till 11am today and found my taskbar full:

Apparently Live Messenger thinks it is so important that I can’t connect that it needs to keep telling me.  Or at least telling the task bar.  This is a bug.

Live Messenger Team:  Please fix this.  I’m sure several of your myriad guidelines for Windows developers eschews any and all of this behavior. 

I was familiar with SSDS before I knew anything about Red Dog Storage Azure Storage.  When I found out about the latter, my initial concern was that Microsoft would confuse developers by offering two overlapping services. Such overlap isn’t too surprising considering that these two projects came out of competing parts of Microsoft.  At the time, there was a pretty consistent message that SDS would someday support relational operations, but to me that meant they should  hold off on SDS until that day came.

Microsoft often offers multiple technologies to solve specific problems — often this is a result of legacy technologies — in this case it seemed a shame to start off with such overlap.

Because of all this, I am very happy to see this clear differentiation between the Azure and SDS services.  This is a good decision for Microsoft, Microsoft developers, and given the roadmap for SDS, an excellent decision for Microsoft’s enterprise customers.

If you aren’t familiar with it, NDepend is a static analysis tool that allows you to dive deeply into your code base.   Its feature list is truly truly impressive. Here are a few things that stand out for me:

• The Code Query Language (CQL) allowing SQL-like queries of your code base, with a set of pre-canned queries targeting code quality, design, naming conventions, and much more.
• A visual tool with dependency graphs and matrices and more
• Ability to compare between different runs of the same project – at an incredible level.
• Of course, A command-line tool for incorporation into your build process.

I loaded up the Digipede Agent assemblies and – kind of like a kid in a candy store – I found myself heading off in 10 different directions at once.  My thought process went kind of like this:

• Wow, look at these matrices and graphs, and all these CQL queries. 
• This is really cool to have this level of information available with build integration!
• Wow, there sure are a lot of warnings here. 
• Let’s fix them!  Wait, let’s prioritize them and customize them and, etc.

(The experience reminds me of when I started using FxCop.)

I highly recommend this software . . .

1. To help you make informed refactoring decisions; and
2. to add design and code-quality criteria (and enforcement) into your build process.

So, why do I say this is a “must-have” tool?  Because code quality is not a nice-to-have.  Quality reduces maintenance and support costs and allows you to spend your time and money on more profitable endeavors.

Go buy it.

BTW: I would like to post some of the results I’ve gotten with the Digipede Agent, but I’m not ready to share that yet. 

Pro:

1. Quality of picture and sound are pretty good (haven’t tested it much).
2. Fast startup.
3. It mounts as a drive.
4. Movies can be viewed and the drive managed without running the Vado Central software.
5. The software / codec are installed on the drive making it easy to plug it into any computer and use it without going through a painful installation step.

Con:

1. The battery door is pretty flimsy. One of the benefits of this device is supposed to be its replaceable battery, but I wouldn’t trust this door to last for a long with frequent battery changes.
2. The battery door doesn’t actually close fully with the battery inside. This problem is solved by putting on the included skin.
3. Like the Flip, this has an extendable full-sized USB adapter. While I understand the benefits of this approach, I would much rather they used a mini USB port instead.  This would allow for either a smaller device – or my preference – room for a stereo microphone input and output.
4. Documentation is bad.  Normally I don’t care about that, but there is no information about the different quality settings.  This can be easily found online, but it should be in the manual.

Bonus Con:

1. I bought the Creative branded waterproof pouch assuming it would be sized for the Vado.  While it will work, it could be much smaller.  The Vado swims in it. With the skin on the Vado it keeps stationary inside of the pouch.
   

So far I’m happy with it.

The old terms are back in effect: delete your account and so goes your data.

A minor success for users everywhere - even those who don’t think this stuff matters.

The change?  Before Facebook relinquished their rights to your data if you deleted your account.  Now they don’t. 

I don’t have a problem with this new policy.  I do have a problem with the new part.

Of course, I’m not arguing whether Facebook can legally make this change, but it does violate their user contract.  I’m not talking about a legal TOS, but of an understanding with their users.  What is the problem?

• Facebook has just asserted ownership to something that they didn’t claim ownership to before.  And this isn’t future data, this is past data.  Data you already contributed to Facebook with an understanding that they wouldn’t keep it.

This is another example of what I call the user-beware contract – where the TOS can change at any time without notification. 

So, what is the user-aware way to make such a change?

• Maintain their old policy for data in Facebook before the change.  This bifurcates user data between before and after the policy. Delete your account?  Old data goes away, new data does not.

OK, but this is still a user-beware contract.  What else should they do?

• Require users to opt-in to the new policy.  If they opt out, either delete them or let them continue the old policy.

I’m sure Facebookians (and any one hosting a large service) is rolling their eyes at this point.  But just because being user-aware is inconvenient doesn’t make it infeasible.

And a shout out to Ned Sykes for prompting this post: no, I’m not concerned about Facebook stealing my tweets, but as a voice in user rights, I am interested in promoting TOS that are pro user.

BTW: The user-beware/user-aware terms are defined in my post User Contracts - Part II: User Beware.

Well… this is embarrassing. We just released an update that’s newer than 14.0.8050.1202. One of the two bugs it fixes, is that our “Check for updates” mechanism broke irrevocably in 14.0.8050.1202 and earlier builds.

If you download the new version from http://download.live.com then “Check for updates” will work again. Sorry for the inconvenience!

So, if you aren’t at least at version 14.0.8064.206, then you should upgrade again.

It looks better and now it renders my blog template correctly.  Maybe there are more features I’m missing.

It is worth upgrading it just to get the “Check for updates” feature.  So you never again have to figure out how to upgrade it (see Jim’s rant here: Sighcrosoft – Why Can’t I Just Love Live Writer Without Confusion?).

Strangely enough, its now easy to upgrade here:  http://download.live.com/writer.

Looks like it will be an interesting program.

If you are going and want to meet up, email me at robert at digipede dot net.

The problem with the law is mainly that the tests are unnecessary in so many cases and the tests themselves are expensive and produce hazardous waste.  Lead in metal toys?  Reasonable to test.  Lead in fabric clothing and linens?  Unreasonable.

A great many businesses are already endangered in this economy, but these extra costs would drive many under.

Good news?

On January 31st, a stay of enforcement was granted (Stay of Enforcement of Testing and Certification Requirements).  This is because the CPSC has recognized that this law is a bad one and that they need to spend the next year providing guidance for a reasonable implementation of the law.

It is clear that the CPSC is attempting to provide clarification to and a reasonable framework around this whole issue.  Further clarification of the current policy was released yesterday.

So, no National Bankruptcy Day?

The answer is mixed.  Much damage has already been done.   The confusion created by this law and the CPSC’s poor handling of it has already hurt many small businesses.  In some cases this is merely the loss of productivity while the founders divert most of their attention to compliance.  In other cases, companies have already chosen to close their doors.

So, today is significant in that the law does go into effect today.  The limits are in place, egregious violators can be punished.  All good, but there has been much collateral damage along the way.

So I’ll say yes, today is still National Bankruptcy Day.  The new, as yet undrafted, policies will go into effect and be enforced on February 10th, 2010 – and they may be just as unreasonable as the current ones.

You can celebrate by giving a giant air-quoted thank you to the politicians and environmentals who rushed this terrible law into being.  I’ll keep my invectives to my self.

The first thing I did get our old Python sample running in IronPython.  This was the first user-contributed sample (thanks to Sean True).  That sample (see it here) uses Python COM libraries to invoke a job with the Digipede Network COM APIs.  This didn’t submit objects, just executed a command-line application.

I’m happy to say that the code required very little modification to run under IronPython.  The only difference is in the syntax of the “import” commands.  Kudos to the IronPython team.

I’ll post the working code once I get a little farther.

Next step: distribute IronPython objects.  Fairly straightforward, but I’ll write a reusable C# Executive to load the IronPython class definition.

I’m working on a project for a customer that includes upgrading a VC++ 6.0 DLL to VC++ 2008.  There are some good resources on MSDN for breaking changes (for a starting point, see http://msdn.microsoft.com/en-us/library/bb531344.aspx).  Of course, I didn’t start there, I started by just trying it.  Pretty quickly I came across some of the breaking changes related to STL (specifically that iterators cannot be treated as pointers).

But I also had a pretty bewildering problem.  Hundreds of the following errors:

error C2371: 'std::tr1::placeholders::ks' : redefinition; different basic types

The source of the error was C:\Program Files\Microsoft Visual Studio 9.0\VC\include\xxbind1, line 298. Looking there, I found:

    // PLACEHOLDER ARGUMENTS
        namespace placeholders {    // placeholders
static _Ph<_NARGS> _CLASS_NAME(_);
        }    // namespace placeholders

Doing some investigation I found that xxbind1 is related to the TR1 extensions to the Standard Library.  The errors didn’t help me figure out where my code was including this file, so using the process of elimination (and the trusty binary search), I found that excluding the STL map header (i.e., <map>) solved this problem.  Not an option, though, since the code in question relies on stl::map.

So, I tried several other avenues to figure out what I was doing wrong.  While scouring all the application header files leading up to this include, I found the following line:

#define _    ks._


I thought, “wow, that looks awfully general for a preprocessor constant in the main internal header file”.  Going back to the code in xxbind1, now that error makes sense.

Removed it, and everything is fine. 

Arbitrary string-replacement before compilation is a powerful thing, but it can allow seemingly unrelated and unintended havoc.

H.R. 4040, the Consumer Product Safety Improvement Act (CPSIA) was signed into law late 2008.  The intention is to protect children by outlawing the sale of children’s products containing lead and other harmful substances. 

Sounds good, right?

Unfortunately, the law is going to force a lot of small companies selling safe products to shutter on or before February 10th.  Welcome to National Bankruptcy Day.  From the site:

February 10, 2009 untold numbers of children’s products manufacturers and retailers will be closing their doors.

The problem is that the law was put together very hastily and has a completely infeasible implementation timeframe.  Oh yeah, and requires unreasonable testing to boot:

Not enough time:  The law was signed in August and goes into effect on February 10th.  So every kids product sold in the US has to be tested in the next 52 days?  That is impossible.

Unreasonable requirements:  Every component of every SKU has to be tested separately even if that component is used in multiple products.  For example, imagine a tag or button that is shared across ten SKUs.  That same button has to be tested ten times. 

Unreasonable cost:  Each test is expensive.  For example, one company has quoted each test at $500.  Continuing with the above example, imagine your ten SKUs with the buttons have 4 other components.  You would have to pay 10 x 5 x 500 = $25,000 to a testing company.

No allowance for existing tests:  Some companies have already had their products tested for harmful substances (e.g., kukunest is Oko-Tex certified).  That should count.

I am all for safe products and accountability, but this is really a draconian measure.  It will punish companies and people with an unreasonable set of rules and penalties. 

If this doesn’t get fixed soon, companies will go out of business, founders and innovators will lose their dreams.  Innovation in children’s products will be all but dead here in the United States.

It will happen unless the US Consumer Product Safety Commission and the US legislature do something to stop it. 

Call to action

Sign this petition: http://www.ipetitions.com/petition/economicimpactsofCPSIA/index.html

Contact your Congressperson and ask them to fix H.R.4040 now, before more damage is one.

Disclaimer:  Do I have a vested interest in this issue?  Definitely.  My wife is the co-founder of kukunest, a company that designs and sells children’s bedding and other products.  I’m a parent and don’t want to see our schools unable to get the products they need because the suppliers dry up.  And I’m an entrepreneur believing that reasonable laws can protect people and protect innovation and business interests.