Is a Windows desktop an expensive and insecure liability that you’d like to get under control, or a flexible and powerful tool that lets your employees work, play and be productive? Martin Banks reported recently on an insurance company who rolled out two whole new overseas offices without leaving the office by shipping out thin clients. You can’t complain about the efficiency for the IT team but I wondered about the difference for the end users. Did they have as much freedom and flexibility - in terms of trying new things or in terms of being allowed the level of personal use people expect (in the UK at least)?

If they can’t, some IT teams might be quite happy, Martin told me. Flexibility and personal use  can be” `danger areas’ seen by IT managements, and if they can bring desktops under better management and control, stopping personal use and cutting the risks of virii infestation etc the better they will like it.”

I asked Microsoft the same question a few years ago, in a discussion about the parental controls and user account protection: shouldn’t companies go all the way and only run white-listed applications and only allow access to whitelisted Web sites and block personal devices… The immediate and forceful reply was that this would be missing the point of the desktop, where anything from a relaxing game of solitaire to experimenting with graphics software could unleash the creativity of your staff. And the point of the PC over the dumb terminal or the Web browser as OS is the range of software you can get, the ease with which you can run it and the way that no matter how obscure the way you fancy spending half an hour, someone will have written some software to help you do it.

You can only concentrate for 20 minutes at a time, they used to tell me at school (in the middle of a 45 minute lesson, ironically enough; I did have one lecturer at Oxford who took a five minute break in the middle of a lecture to chat about his kitten on the same principle). Whether it’s relaxing by watching a Star Trek mashup to the tune of White Rabbit on YouTube, losing a hand of Spider Solitaire or spending ten minutes doing your online banking rather than an hour in the queue at the hole in the wall they call a bank branch these days, personal use makes the users happier. Personally, I think if personal use is excessive it’s time for a chat with your manager rather than a visit from the IT team.

The assumption with thin client is that you can’t do that any more. It’s my own assumption usually. But when Aspen talked about the way they’re using thin clients for branch offices at VMWorld today, they had a very different emphasis. Yes, they liked just shipping out thin clients to the office and letting the brokers plug in their keyboards and screens and get on with it without any IT setup time. Yes, they like not having to send technology support out to the office - and not worrying about servers left under the desk in an office that’s not locked until the cleaner goes home.

But a couple of days after they sent the first batch of thin clients out to the office, they had to replace them with Wyse models because they’re better at streaming AV. The brokers need access to the BBC and Bloomberg and videos of CEOs reading out company statements - and YouTube (although it’s not clear if that’s business or pleasure). Aspen supports any USB device the brokers want to plug in from printers to iPods. They chose their current system over Citrix so they could have Vista and XP desktops and a full range of software. After all, the brokers are probably well paid and in demand: tie down their IT system until they’re tripping over it and you’ll lose more money than you save on not sending people out for a desk-side support visit.

That’s good news for the users and thin client obviously works well for Aspen. But it means that picking thin client over a desktop PC might not be the decision you originally thought you’d be making.

-Mary

The UK government wants ISPs to control the traffic of illegal file sharing through their networks. If they don’t, the government is planning to introduce legislation to ensure they do. I’m left stuck with an image of King Canute trying to hold back the tide. Failed business models don’t need to be propped up with legislation…

So why isn’t it going to work?

First, some basic numbers. The BBC article on the government proposals suggests that there are 6 million broadband accounts downloading illegal files every year. That’s a hefty proportion of the UK’s 15 million or so broadband users. While broadband analysts Point-Topic predict that number to grow significantly over the next few years, it’s reasonable to expect file downloading to grow at a similar (or, More likely, faster) rate.

Then there’s the problem of identifying the traffic. While it’s possible to roughly identify the application sending each packet, it’s impossible to say exactly what it’s being used for. There’s no point in just labeling BitTorrent connections as illegal downloads, especially as many open source projects use it as a tool for reducing the load on their servers and for giving users faster downloads. There’s also the issue of identifying BitTorrent connections, as many clients connect on random ports and encrypt their connections. To a packet monitor a BitTorrent client delivering a copy of Open Office looks much the same as the latest episode of Doctor Who.

Traffic volume isn’t a signifier, either. That big download could be an MSDN file transfer of an ISO of Visual Studio or Windows Server 2008. It could even be someone using a cloud storage service like Mozy to upload several gigabytes of photographs to a secure online backup service. Or perhaps it’s someone using 4OD or the BBC iPlayer to download a TV programme they missed. With consumer DSL finding a business role as a tool for connecting branch offices it could even be an estate agency updating its database (with the myriad digital photos a house sale needs these days), or an insurance broker delivering a batch of scanned and signed forms to head office.

The bottom line is quite simple: it’s virtually impossible for ISPs to economically identify and filter user actions that infringe on copyright files. The cost of implementing filters is prohibitive (look at how long Google took to even start filtering YouTube), and the time needed to identify exactly which users do what over the network will detract from actually managing and running a commercial network.

There’s also another part to this story.

Running a consumer ISP is hard enough without having to cope with the additional demands of regulators. Customers are on the phone 24×7 demanding service levels that any business IT department would find impossible to implement. An increasing range of IP connected applications and services are stretching thin budgets to the very limit, as network engineers try to emulate a Star Trek chief engineer putting the Enterprise back together with nothing but string and sealing wax.

It’s an issue I’ve had personal experience with, as I used to run the technical side of UK Online - and that was back when dial-up connections were the norm. If it was virtually impossible then to manage usage, what’s it like for today’s ISPs that see an explosion of protocols and packets across their fragile networks.

Illegal filesharing is the least of an ISPs problems. Legal file sharing is a much bigger problem, as the protocols used by services like the iPlayer are inefficient compared to BitTorrent, and the expected traffic volumes are more than likely to overstress existing interconnect and backhaul bandwidth. It’s probably fair to say that ISPs are now finding themselves squeezed from both ends.

If bandwidth is an issue, then ISPs will find ways of controlling it. Plusnet is traffic shaping to reduce network load at peak times. It also produces graphs to show just what type of traffic is used - and when.

Plusnet also produces indivudual reports for each user to show what traffic they were generating. It’s an approach that helps users schedule their own downloads to appropriate times - reducing overall load, and letting users choose their own quality of service.

If  ISPs aren’t going to feel under the pressure, then may be this is the type of approach that’s needed.

–Simon

Actually, if you’re read your Dilbert, you’ll know that’s not true and it’s really photocopiers; employees come in around number nine on the list (and a copier manufacturer once sent me the statistics to back it up).

But number one or number nine, your employees are the source of your business’s knowledge and ability; you may think you have business processes but it’s your employees who actually get things done (often by entirely different methods). So what technology are you using to manage that asset and how much more can you get out of it?

HR isn’t just there for the nasty things in life; if you use it properly, HR could be a repository for useful information about the skills in your organization, the candidates who turned you down but might be worth approaching again in a year’s time - and when the most people will be out of the office, if you’re trying to pick a good time for a major network upgrade. The HR team knows when new employees will start work; you could get the manager to ask the IT team to order them a PC while the facilities team books a desk and a phone line - or you could have a system that does it automatically when HR enters the day they’ll start work.

You need to know when people actually make it to work, when they’re on holiday, ill or just AWOL. Do you let them enter that into the system themselves, or have them fill out a form, show it to their manager and send it on to a third person to put into the system? Sounds like an obvious choice but according to Chris Berry, MD of HR automation specialists Computers in Personnel, efficiency isn’t what a lot of companies are thinking about. At one large company that he’s too polite to name, when Berry suggested an automated system employees could use directly, the head of HR told him they couldn’t consider it; after all, what would the 50-strong admin team do then?

You can’t leave everything up to employees; there have to be checks and approvals for some processes. But there are plenty of approvals that can be automated as part of a workflow without taking up two people’s time. If I’m putting in an order through a purchasing system and it’s under my sign-off level then it shouldn’t need to be countersigned. If I want to update my bank details, I shouldn’t have to mail it to someone who prints of the email, hands it to an admin and has it typed in ‘to make sure it’s right’. I’m motivated to give you the right details in the first place, because I want to get paid, and if I get them wrong, I’m motivated to come back and correct them - or I don’t get my money. And if I’m using a self-service system, the form can have validation built in so I have to type in a sort code with six digits; you can’t do that in email!

It’s joined up business; not only do you save on admin time and get more of your data right first time, you’re bringing another layer of information into systems where you can analyse it. Usually we think about upgrading technology to make a server or application run better rather than because it’s slowing individual employees down; but if you could see that you get more transactions through a  server when half your team is away on holiday, you might have some different questions about load balancing.

There are the nasty things to take care of too; from compliance audits to spotting office bullying through odd patterns of who’s taking days off, to outright fraud - CIP has found more than one ghost employee, on the books just for payday every month.  And there’s one publicly listed company with tens of thousands of employees that’s started using CIP’s software and noticed that every single senior manager is within five years of retiring. That has to count as important business intelligence…

-Mary

My phone knows where I am, and when I flew to Geneva the other week it knew what time it was; the operator pushed a time signal and Windows Mobile 6 happily picked it up. It confused me when I took the phone out to change the time - but it also meant the appointment with the contact number for the taxi driver was up on screen where I needed it. I connected my PC to the Orange World Wi-Fi in the hotel (at the fifth time of asking; if you’re using a mix of numbers and letters as your username and password, please use a font that allows the user to distinguish 6 and G ). My PC sat there stubbornly believing it was on UK time, even though it had a French IP number.

I’m not expecting every PC to have a GPS in, and it doesn’t need to. Never mind battery life, it’s useless inside anything bigger than a garden shed and even in a city canyon it’s impractical; it took my O2 XDA Stellar 15 minutes to get a GPS fix in Covent Garden this week. What I’m after is a utility that uses the location services like Spotigo, Aruba, Navizon, PlaceSite, Skyhook and all the rest that give you location based on your IP address/what wireless access points you can see and when it gets a location that’s different from the time zone Windows is set to, up pops a prompt asking if you want to change it. If you want to be all social networking about it, the utility could upload my location to services like Facebook - or preferably just my timezone, as I’m sure burglars read Facebook too. I could have a widget in the Sidebar showing who’s in the same timezone as me or get an alert if someone I know is in the next street.

I’ve used Navizon on Windows Mobile for the last year to get locations and I like it but the desktop version is a Java applet and although the API supports time as part of the location info I haven’t found a timezone utility for it.

Skyhook’s Loki will do the locating and publishing bit. It’s pretty good at locating too; this is the service used by Google Maps on Windows Mobile and the iPhone and it knows where we live. Skyhook can use a combination of GPS, IP address, Wi-Fi and cell tower to cope with a range of environments.

Navizon uses user-contributed data for Wi-FI and cell tower and is either very accurate or about 2 miles out; Loki (and Google Maps Mobile) are either very accurate or not working at all.

Loki s obsessed with search; that’s because ads you click on make money. Personally, results in the same town as me may or may not be more relevant to be depending on how far ahead I’m planning and I don’t actually want any more browser plugins, thank you. But digging through the options - yes, it will change my timezone for me, or ask if I want to in case it’s wrong.

This would be a good service for tools like Xobni to use; this handy Outlook plugin shows a ‘heat map’ of the times of day a particular person sends and replies to email. That’s pretty useful already - it tells you that you have a much better chance of getting a reply from me between 10.30 am and 7pm or between 11pm and 1am than at any other time. Assuming I’m in the office; the location timezone service could tell you if I’m in California - and if Xobni was really smart and I said it was OK for you to know where I am (cue my usual call for an identity abstraction layer for the Internet), it could shift the heat map to California time. Or better still, it could calculate a different heat map for when I’m in California, when you’ll reach me between 9am and 11am, 2pm and 6pm and 9pm to midnight most days.

At the moment you can look at my Dopplr trips, or my Facebook status, or my most recent personal blog post or the last photo I posted on flickr to work out where I might be - if I’ve remembered to update them and you remember to check them (a friend assumed I’d be in Barcelona for Mobile World Congress this week, and ended up having a night in instead of coming round for dinner). That’s both of us doing extra work that the computer should be taking care of and I’m sure that’s the wrong way round. 

There’s two halves to this. One is that location is a really useful service (see my 2008 Technology Resolutions), especially as more of us work from home, travel around more and run out of time to arrange meetings with friends. And that’s the really big thing. I want computers to start saving me time and getting more done for me, not by making it faster to get my accounts done or by letting me try 90 versions of my Web site in the time it used to take to write one, but by working out the context and giving me opportunities. If my To Do list says I need to get something from the Lurgashall Winery for a friend and I get a message from a friend in Billingshurst needing help with something and a mail from a client in Horsham wanting to talk about work, having my PC suggest that I’m in Guildford on Monday is handy (and we think it’s why Microsoft wants Yahoo!); having it know I’m actually in Guildford today even though I didn’t update my calendar and give me an itinerary for the afternoon is even more useful. And it’s the computer doing the running around, not me. For that, I’ll put up with another browser plugin.
-Mary

Mary and I are a small business, and we rely on our workhorse of a server. Currently running Microsoft’s Small Business Server, it sits in the corner of the office and looks after our mail and our files. Sadly, though, it’s getting close to retirement. A solitary Athlon isn’t really up to today’s workloads, and many of the features we want from Exchange are only in the latest release - which is 64-bit only.

Our new server has arrived, and it’s surprising what a few years have done to the SMB server market. We’d had to build our original server ourselves, but this time we’ve ended up with a dual-core Xeon system from HP. I’d been delaying purchasing a new server as I was expecting to pay through the nose for my hardware, but when I started looking at current prices I found I couldn’t have been more wrong.

That HP server? With 512MB of RAM and a 160GB hard drive, it cost us less than £180. I was able to buy 2GB of RAM for around £50, and two 500GB hard drives for under £100 all-in. That’s around £320 for a machine that’ll meet most of our business needs for the next four or five years. Fititing it all together was easy enough, and the server was powered up and ready to go in under an hour. Once it’s all configured I’ll be adding an eSATA card for speedy backups.

Breaking it down a bit, it all means that our server will cost us around £80 a year, before software and maintenance costs. Sure, it’s not a 1U rack mounted system, or a set of blades, but it’s a sturdy well engineered box with a processor that’s hefty enough for most needs.

Hardware is the easy bit. Software is a lot harder. I’ve now got to plan just how I’ll migrate files and mailboxes to the new machine - and how I’ll handle decommissioning the old box. If anyone’s got any good tips, I’ll be glad to hear them!

 –Simon

If you get a smartphone from work, you may not get exactly what you want, but at least the IT team will set it up for you and deal with things if you lose it.  Buy your own and you’re on your own - usually. Buy an i-mate Windows Mobile phone - I quite like the look of the new Ultimate 8502 as a replacement for my trusty HTC Excalibur as it adds HSDPA/HSUPA and GPS to the QWERTY smartphone - and you get I-Q Services.
 
I-Q is a free service on the Club i-mate Web site that lets you configure, manage and brick your phone over the air. Unlike some of the meaner handset manufacturers, i-mate ponies up the licence cost for the full apps for Mobile Office, but if you want to add Opera as your browser or a different IM tool you can choose what you want on the site and have it sent over the air to your phone. You can even put the apps into the ROM so they’re not taking up extra space. And if you lose the phone, you can go onto the site and lock it so it can’t be used; if you’re sure it’s been stolen you can shut it down permanently. The feature for that is in ROM so the thief can’t turn it off and they can’t pull data off by turning off the radio and connecting it to their PC - as long as their PC is online, the phone will get the kill command over the Internet. And if the thief tries installing a new ROM image on the phone to get it working again, they’ll be asked for a 15 digit code, without which the boot loader won’t load the boot image.

This is the kind of service a big customer might get from an operator; i-mate is running it for free for every user. CEO Jim Morrison thinks it will sell more phones; if you have the choice of two handsets with the same features and one of them comes with a free management service, I know which one most IT departments will pick.

There’s another security feature that i-mate hasn’t turned on yet. Some of the new models include GPS - Morrison says “every high end phone going forward will probably have GPS in”, because the $12 that it costs to put the chip in is peanuts compared to the value people put on GPS in a phone. And when you have GPS, then you have a phone that knows where it is and can report back. To avoid Big Brother spying, the tracking will only work when the phone has already been locked. The IT team can’t keep a sneaky eye on who is really with a customer and who is down the pub, because the first thing you’ll do if your phone suddenly locks up is contact the IT Team. Eventually, Morrison hopes, thieves would get the message that i-mate devices just aren’t worth stealing…

Everyone has to compare every smartphone to the iPhone (I think it’s an EU regulation or something). I’m fine with that as long as the comparison covers email search (BlackBerry and Windows Mobile 6 tied for equal honours, iPhone nowhere) as well as the stunning interface and graphics on the iPhone. NVIDIA’s Michael Rayfield thinks it’s fair. “We talk a lot about the iPhone not because of the device but because it redefined what’s good enough. It’s a computer that can make a phone call. It’s got a robust OS -  it’s got OS X, the whole stinking thing - and that’s necessary, we think, to have that going forward.”

A more powerful OS needs a more powerful processor, better graphics - and the same or better battery life as today’s phones. NVIDIA set the Portal Player team it bought to work on an application processor for phones (think of it as a very low power graphics card) and 800 man years later it’s announcing the APX 2500, a 750MHz ARM 11 processor that does hardware acceleration, 720p HD, transparency and 3D in an ultra low power version of the GeFORCE, so you can have the kind of user interface you get in Mac OS X and Vista (plus HD video playback on screen or through an HDMI connector onto a real screen). Rayfield says a phone using the processor should be able to play 10 hours of HD video or 100 hours of MP3s, so you can get on a plane to the US, watch a movie, listen to music the rest of the way and still have half your battery left for making calls when you land.

NVIDIA is pushing the processor for Windows Mobile phones first and then Linux and Symbian devices. It’s going to take a while to convince manufacturers and who then have to build phones and write applications and integrate NVIDIA’s UI or build their own (and this is far more complex programming than for current mobile phones). Some personal media players and navigation devices will come out using it at the end of this year and then some time in 2009, we could start seeing some really spiffy phones. At which point, you’ll definitely want them to be thief-proof.
-Mary

The barracuda is the wolf of the sea, a slim silver dart that hunts in deadly packs. It’s perhaps not surprising that NVIDIA has taken part of its name for its GPU-based supercomputing tools.

On a recent trip to the US, Mary and I met up with some of the folk behind CUDA at NVIDIA’s Sunnyvale headquarters. It was a fascinating conversation - if only because I used to write scientific computing software, and something like CUDA would have sped up my work massively. When a problem takes days to solve, something using something like CUDA to accelerate processing makes a lot of sense.

Prior to CUDA, NVIDIA had tried to use GPUs for compute, but had run into architectural problems. Things changed with their series 8 GPU, which was very different to anything they’d built before, being designed for compute as well as graphics. That’s lead to some tradeoffs - there’s silicon on the GPUs that’s unused when it’s used as an accelerator (and vice versa). However NVIDIA makes so many chips, there’s not really any financial issue, it all comes out of the economies of scale.

CUDA is more than just a set of chips - it’s a language framework for working with GPUs, that can andle both sequential and parallel code together. Developers don’t need to learn anything you, and the framework gives programmers explicit - and simple - interfaces for running parallel code on NVIDIAs GPUs. There is a long term goal of providing tools for automating parallelism, but at this point you still need to work out what code can be parallelised yourself. The result is code that’s very simple with much less code, as CUDA handles repetitive calculations for you.

Simplicity comes from the hardware as well, as it manages threads for you. All you need to do is define the tasks the GPU will handle, and manage their interactions. The GPU then runs the calculations over the data, with groups of processors on different functions at the same time. As RAM is directly attached to the GPU there’s no need to use the PC’s own memory for caching data.

The numbers coming out of CUDA are impressive. Working with the VMD/NAMD molecular dynamics tools researchers at the University of Illinois have seen a 240X speed-up in the VMD ion placement tool, and an 8 to 12X speed up in NAMD. With an eye on greener computing, they’re also finding that CUDA gives them 1W/Gflop!

If you want this sort of power for your applications (and it’s remarkably suitable for large financial applications) you can by NVIDIA’s Tesla systems. There are work station versions, along with deskside offload processors. However the version we were most impressed with comes as a 1U rack mount unit, containing 4 GPUs. Connected to a PC or a server via 5 Gbps PCI-Express connections this is the way to give your data centre applications a significant speed up, with significantly lower power requirements.

While Tesla may not yet meet NVIDIA’s aim of providing a Teraflop in a 1U unit, it certainly speeds things up. Oxford University researchers have used it to get a 149X speed up LIBOR risk analysis for an 89X improvement on performance/Watt. That’s a good deal in anyone’s book - especially if you’re working with today’s fractious financial markets.

Add one to my list for the IT Santa!

–Simon

Anyone who’s listened to me rant over the last couple of years will have heard me say that I expect Microsoft to takeover Yahoo!. With Google’s dominance over search, and Yahoo!’s success at what Microsoft wants to build into Live, there’s a certain logic to a merger of the two businesses. Yahoo needs the R&D boost that Microsoft can give it, and Microsoft needs the online presence of Yahoo!.

I wasn’t surprised to see that Microsoft has made a formal offer to Yahoo!, offering $44.6 billion for the company.

Microsoft’s been playing nice with Yahoo! for some time. It’s Windows Live Photo Gallery handles uploads to Flick, andWindows Live Messenger can talk to Y! users. The love goes both ways too, as Y! is one of the first applications to really take advantage of the Vista UI enhancements.

Reading Steve Ballmer’s letter to Yahoo! this morning I noticed a couple of quotes.

The first is the one everyone’s expecting. Microsoft needs Yahoo! to compete with Google:

While online advertising growth continues, there are significant benefits of scale in advertising platform economics, in capital costs for search index build-out, and in research and development, making this a time of industry consolidation and convergence. Today, the market is increasingly dominated by one player who is consolidating its dominance through acquisition. Together, Microsoft and Yahoo! can offer a credible alternative for consumers, advertisers, and publishers. Synergies of this combination fall into four areas:

Scale economics: This combination enables synergies related to scale economics of the advertising platform where today there is only one competitor at scale. This includes synergies across both search and non-search related advertising that will strengthen the value proposition to both advertisers and publishers. Additionally, the combination allows us to consolidate capital spending.

Microsoft has been talking about online advertising as a key play for some time. There’s a lot of advertising money out there, and even if you add in the efficiencies of online over other mechanisms, not enough if it is going to the online side. Google may have a lead at the moment, but it’s one that can be eroded. Besides, there’s enough money out there for everyone to do very well thank you.

The second quote is, for me, the more interesting:

 Emerging user experiences: Our combined ability to focus engineering resources that drive innovation in emerging scenarios such as video, mobile services, online commerce, social media, and social platforms is greatly enhanced.

The industry is going through a big change right now, and I thonk it’s one that Microsoft has its finger on. It’s not the move from offline to online - it’s the fact that computing power now has finally reached the point where the boxes on our desks can finally be really useful.

Sure, they can add up and act as glorified type writers. That’s not what we signed up to when we got on the IT train. Management information systems were part of the story, but the real game changer is getting applications to understand context.

Context is a big problem. It’s not just the fact that I’m online now, it’s where I am, what I’m doing, what my diary says, what my friends and colleagues are doing, what my boss wants me to do - it’s everything about me and my world filtered and annotated to help me make decisions. Think of it as SAS-style business analytics about yourself, using self generated social networks from tools like Xobni.

That’s where Yahoo! comes in. It’s been doing a lot of work on tagging and on determining meaning in tags. You’v probably seen pictures of Jerry Yang’s demo at CES, of a map of Las Vegas overlaid with Flickr and Upcoming tags. Yahoo!’s been showing that demo for a while now, and it’s part of the context story: building applications that understand me and my needs.

The endgame is Clippy done right (and on steroids). Instead of an annoying “I can see you’re writing a letter” your copy of Outlook says “I can see you’re going to Las Vegas for a conference. Do wish to contact your colleagues who will be there? Susan has free time in here diary in Wednesday evening. She’s vegan, so here’s a list of suitable restaurants. I’ve also noticed that Jet Blue has a special air fare for the days you want to travel and that will give you enough points for next holiday flight. There’s also a room in your preferred hotel. You might want to pack sun screen - it’s going to be hot.”

Your computer knows a lot about you. Perhaps it’s time for it to  start putting it to use. Microhoo! could just be the way it starts to do it…