USB flash drives: a modern design canvas

USB flash drives were invented around 12 years ago. They soon became commonplace, so designers differentiate with creative designs. I have a drawer full of them and have picked out some that caught my eye.

I like the understated elegance of this Adobe stick.

image

though for elegance perhaps this Kingston is the winner. Paperclip included so you get the scale:

image

This Huawei stick pays homage to Rubik’s Cube:

image

This Google man is a favourite:

image

though for the full effect you have to plug him in:

image

The designer of this Asus stick plays on the fact that they are sometimes called USB keys:

image

This one from Supertooth is a fake music player or something:

image

Marley goes for the natural wood effect of course:

image

Finally a reminder of where we started. I am not sure of the date of this stick but I have not attended a Borland event for many years:

image

It has an LED that lights when plugged in. But the real shocker is the size, shown on the back along with a rather obscure warning:

image

Still, bearing in mind that a floppy disk was not normally bigger than 1.44MB, 16MB is not to be sniffed at.

I also suggest that the era of USB flash drives will soon pass. Apple does not support USB storage in iOS, other than to a limited extent for cameras, and just as CDs gave way to USB drives, the USB devices will be replaced by wireless transfer, either locally or via the internet. Some press releases now arrive with links to Dropbox folders. How sensible.

Moving a database from on-premise SQL Server to SQL Azure: some hassle

I am impressed with the new Windows Azure platform, but when I moved a simple app from my local machine to Azure I had some hassle copying the SQL Server database.

The good news is that you can connect to SQL Azure using SQL Server Management studio. You need to do two things. First, check the server location and username. You should already know the password which you set when the database was created. You can get this information by going to the Azure portal, selecting the database, and clicking Show connection strings on the dashboard.

Second, open the SQL firewall for the IP number of your client. There is a link for this in the same connection string dialog.

Now you can connect in SQL Server Management Studio. However, you have limited access compared to what you get as an admin on your local SQL Server.

Here is the Tasks menu for an on-premise SQL Server database:

image

and here it is for a SQL Server Azure database:

image

Still, you can start Export Data or Copy Database from your on-premise connection and enter the Azure connection as the target. However, you should create the destination table first, since the Export Data wizard will not recreate indexes. In fact, SQL Azure will reject data imported into a table without at least one clustered index.

I tried to script a table definition and then run it against the SQL Azure database. You can generate the script from the Script Table as menu.

image

However even the simplest table will fail. I got:

Msg 40514, Level 16, State 1, Line 2
‘Filegroup reference and partitioning scheme’ is not supported in this version of SQL Server.

when attempting to run the script on SQL Azure.

The explanation is here.

Windows Azure SQL Database supports a subset of the Transact-SQL language. You must modify the generated script to only include supported Transact-SQL statements before you deploy the database to SQL Database.

Fortunately there is an easier way. Right-click the table and choose Generate Scripts. In the wizard, click the Advanced button for Set Scripting Options.

image

Find Script for the database engine type, and choose SQL Azure:

image

You may want to change some of the other options too. This generates a SQL script that works with SQL Azure. Then I was able to use the Export Data wizard using the new table as the target.  If you use Identity columns, don’t forget Enable identity insert in Edit Mappings.

image

Farewell to Microsoft Small Business Server

Microsoft has announced pricing and licensing for Windows Server 2012. A dry topic perhaps; but one which confirms the end of a product with which I am perhaps too familiar: Small Business Server. It is spelt out in the FAQ:

Q33. Will there be a next version of Windows Small Business Server 2011 Standard?

No. Windows Small Business Server 2011 Standard, which includes Exchange Server and Windows server component products, will be the final such Windows Server offering. This change is in response to small business market trends and behavior. The small business computing trends are moving in the direction of cloud computing for applications and services such as email, online back-up and line-of-business tools.

The next question confirms that there will not be a new edition of Small Business Server 2011 Premium either. The official replacement is Windows Server 2012 Essentials, which is in effect the next version of Small Business Server Essentials. This handles local Active Directory, file sharing, local applications, and a connector to Office 365. However there is a 25 user account limit, whereas SBS standard supported up to 75 users, so there will be some businesses who are now forced to choose between moving to Windows Server Standard, or ditching the local server completely (which is often impractical).

image

Microsoft is pinning the reason on cloud computing, which makes some sense. Now and again I am asked by small businesses what sort of technology they should adopt; and my answer in general is to point them at either Microsoft Office 365 or Google Apps.

It is not quite clear-cut. A Small Business Server can theoretically work out cheaper, if you presume that it will not require any external maintenance. That is rarely the case though, and for most people the cloud-hosted option will be both cheaper and less troublesome.

What if you do need on-premise Active Directory, Exchange and SharePoint, which are the core components of SBS? Technically, there are in my opinion better ways to do this than with SBS. While SBS has always been excellent value for money, it is over-complex because it crams onto one box applications which are designed to run on separate boxes. It does work, but if anything goes wrong it is actually harder to troubleshoot than when you have separate servers. I prefer to see one Hyper-V box with separate Virtual Machines (VMs) for each major function, than SBS running on bare metal. VMs are also more flexible, and easier to restore if the hardware breaks.

Farewell then to SBS. I will remember it with some affection though. Think back to the nineties, when most email was POP3, and most internet was dial-up. People had problems like losing emails, because they had been downloaded to a desktop PC and they were out and about with a laptop. Moving to Microsoft Exchange, for which Outlook is the client, was bliss by comparison. Email synchronised itself to all your PCs, you could work offline, and Outlook for all its faults became a one-stop application for calendar, contacts and messages.

The beauty of SBS was that you could get Exchange along with the benefits of a Windows domain – one central directory of users and the ability to assign permissions to file shares – at a price that was more than reasonable.

I also think of SBS as a reliable product, when correctly installed. When it does go wrong it is often due to users trying to do stuff that does not quite work, or other applications which get installed on the same box, or hardware faults which users have attempted to fix by messing around with Windows, or anti-virus software misbehaving (Sophos! Confess!).

Microsoft is doing the right thing though. The SBS bundle makes little sense today, and if you do still need it, you can stick with the 2011 edition for a few years yet.

Embarcadero adopts open source Clang for future C++ versions

A couple of months ago Embarcadero’s John Ray Thomas published a roadmap for the company’s C++ tools. Coming soon: not only a long-awaited 64-bit compiler for Windows, but also native iOS and Android support. On top of that, there are plans for “the very best in C++11 and C99 language and library compliance in the industry.”

Sounds good; but this forthcoming upgrade is not quite what it seems. I spoke to technical evangelist David Intersimone about the changes. The company is adopting Clang, an open source project which creates a C/C++/Objective C front-end for the LLVM compiler. “We’re integrating all that into our IDE,” said Intersimone. “We’re also going to be using that same toolchain world with our Delphi compiler as well.”

“We have analysed what to do about C++. Our compilers both for Delphi and C++ have been around for a lot of years, and over time it just got harder and harder to add new capabilities to make programming simpler and also to add power and richness to the languages. In C++ in particular the language continues to be updated, with now C++ 11. So we made the decision to use Clang and LLVM on the C++ side, for the 64-bit compiler. We’re going to keep our existing compiler for 32-bit Windows for now, and then eventually replace that.

“For Delphi we’re still using our existing compiler to do some of the work, but we’ve been working on a next-generation compiler for Delphi and that is still in the works. For a while we’ll have two compilers, the existing architecture compilers, and then new compilers.”

Embarcadero has its own C++ extensions to support component development, and is working on adding them to Clang. “We’re leveraging and extending the Clang compiler with the property-method-event extensions that we added to our own C++ compiler.”

It is a sad moment in some ways, bearing in mind the long history of what was once the Borland C++ compiler. Equally, it is a sensible move. Intersimone said that the work of keeping up with the evolving C++ specification was disproportionate to the benefits. “It’s a monster language, with a lot of power and a lot of complexity. It made perfect sense to fit our extensions [to Clang] versus building a compiler from scratch and having to continue to track the language into the future.”

Embarcadero can focus instead on its IDE and tools, and on the frameworks for Windows and for cross-platform.

Look out for a more detailed interview with David Intersimone in a future article for Hardcopy.

aQuantive may be Microsoft’s biggest acquisition failure. Have there been good ones? A look back.

Today’s news that Microsoft  is writing off $6.2 billion from the useless acquisition of aQuantive in August 2007 gives me pause for thought.

How bad is this company at acquisitions? Particularly those under CEO Steve Ballmer’s watch. He became CEO in January 2000.

image

Microsoft acquired Danger in February 2008 for $500M. Small relative to the aQuantive acquisition, but how much further money did the company burn transforming Danger from an excellent cloud and mobile company to the group that came up with Kin, the phone withdrawn from the market after just two months on sale? Not to mention the downtime and threatened loss of data suffered by Danger’s online service under Microsoft’s stewardship.

Microsoft attempted to buy Yahoo for $44.6bn in 2008. Yahoo’s executives declined, a move that was (very) bad for Yahoo shareholders but quite possibly right in a business sense; it would not have been a good fit.

Microsoft acquired Groove Networks complete with Notes inventor Ray Ozzie in March 2005. I put this in the disaster category. Groove went nowhere at Microsoft. Ozzie became Chief Software Architect and talked of internet vision but did not deliver. The wretched SharePoint Workspace is apparently based on Groove.

What about the good ones? My view is that Microsoft paid too much for Skype at $8.5 billion but at least it acquired a large number of users and has some chance of enhancing its mobile offerings with Skype integration.

Microsoft acquired Bungie in 2000 and given the success of Halo (without which, maybe, the whole Xbox project might have faltered) we have to count that a success, even though Bungie was spun off back to independence in 2007.

Other notables include Great Plains in December 2000 (now morphed into Dynamics ERP); Connectix in February 2003 which got Microsoft started in virtualization; and Opalis in December 2009 whose software now plays a key role in Microsoft’s System Center 2012 private cloud software.

Winternals in July 2006 was a great acquisition. Microsoft acquired some indispensable Windows troubleshooting tools, and also Mark Russinovich and Bryce Cogwell, able people who I suspect contributed to the transformation of Windows Vista into Windows 7, and in the case of Russinovich, to the technology in Windows Azure which now seems reborn as an excellent cloud platform.

You can see all Microsoft’s completed acquisitions here.

(If the company would like to acquire itwriting.com for a few billion I am willing to talk.)