Blogs

Redmonk

CloudCamp London, Avoiding Monsters

“The idea of cloud computing — designed around an architecture whose natural state is a shared pool outside the enterprise — has gained momentum in recent months as a way to reduce cost and improve IT flexibility. But the use of cloud computing also carries with it security risks, including perils related to compliance, availability, and data integrity.”

- The dangers of cloud computing, Ephraim Schwartz

Something about this piece really got my gander up. Dangers and Perils- Here Be Dragons. Ah yes the beauty of FUD.

here be dragons

For those of you that realise the world isn’t either/or and that we’re going to see a mix of on premise and cloud, and that for all the initial irrational fears there will be riches and assets to be discovered out there. For those that realise if we start with cloud computing we’re not going to sail over the edge of the world, we bring you… Cloud Camp London. Its tomorrow, July 16th, at the Crypt on The Green in Clerkenwell. See you there

the image above comes from Olaus Magnus.

Osmosis by Osmosoft: BT’s Open Source porous membrane

Last Friday I went to my first Osmosoft Show and Tell Session. As is the way of these things, I was more into the chatter at the bar afterwards than the actual event. The bar itself was a metaphor for the change I am about to describe.

The Exchange Bar is.. an old BT telephone exchange. It was used for routing the phone calls leading to millions of Soho assignations. Today its a bar, with reasonable music, Kirin beer on draft, and half decent pizzas. The infrastructure has been transformed, as have the services offered. You can probably see where I am going with this…

BT acquired Osmosoft, the company behind a nifty, open source, Javascript-based tool called Tiddlywiki, last year. Now Osmosoft is emerging as ”the open source arm” of the telco. People from the Osmosoft team develop software, but also help to socialise Open Source, helping the rest of BT to understand issues around intellectual property and so on. What does it mean to consume open source, and perhaps more importantly what are the advantages of contributing code?

Osmosoft is a social object in its own right, acquired by BT to foster a new conversation, both internally and externally. Paul Downey, now a BT “chief web services architect” is known for social objects such as this one. Its worth visiting Flickr to check out the annotated version. So who better to seed BT’s Blue Monster?

The Web is Agreement

Why should BT care about open source? Why does it need a Blue Monster? For a number of reasons, not least the cost-savings available to companies building on open source foundations, and engaging with open source community collaboration models. BT’s job is to build and deliver network services, not infrastructure middleware. It can, and does, spend shareholders money on proprietary software, but it doesn’t always *have* to. Cloud computing is the latest area Osmosoft is looking at - it could spend all of its money on BEA WebLogic and Azul Systems appliances or it could try for more lightweight approaches to cloud provisioning for users. Which is where guys like Dr Pep come in.

I like to point to the Eclipse ecosystem (which used to be the Java IDE market until open source hollowed the market out) for those that haven’t understood this shared IP dynamic yet. Shared open source foundations tend to be stable foundations, and are designed for integration and repackaging.

BT is not reknowned for its speed of innovation, nor its speed of service development, but that is changing these days. One of the BT developers, Dr Pep again, told me that his dad had recently complained we wanted BT to stop innovating, because they have too many new products already. BT is in “many flowers” mode, and people like Paul and the Osmosoft team are going to help seed them.

Osmosoft will help to punch through BT’s semi permeable membrane. It will foster a developer community, particularly in the open source area. It will help BT understand the web. I am certainly going to spend more time with the Osmosoft guys in future. If nothing else they are really good company…  BT’s transformation won’t be immediate, but buying outfits like Osmosoft will certainly help the transition.

OSGi Stackless Component Model meets AJAX, Android, Flex

Great little video here from Eclipsecon 2008 with Cote and friend of Redmonk, Neil Bartlett. In case you haven’t heard of OSGi its time you did - its becoming an increasingly important standard, driving the rise of the Stackless Stack. It ties enterprise and client-side middleware together, with a compelling development model through Eclipse-based IDEs.

5 minute blog: HP gets its research mojo?

I am going to write this in 5 minutes.

I recently visited HP R&D Labs in Bristol, and the trip was worthwhile.

What did I learn?

HP has a notion called Everything As A Service. The idea makes a lot of sense- I tend to call it the Service Mass Convergence; HP’s term is far catchier. Literally anything it seems can be provided as a service, and probably will, especially as become more and more digital. HP actually started on a v similar notion back in 1999 with a play called e-services. It didn’t work out.

HP’s recent record on organic innovation is frankly pretty poor, which the company needs to address. Frankly some of its acquisitions for innovation have also been undercooked (Bluestone is the poster child). So HP needs to get more out of its research labs - a bunch of squeak was never going to drive the bottom line. Frankly this trip the folks at labs came across as far more commercially aware.

Being less academic by being more academic: I was impressed by the new labs director - Prith Banerjee. One of his key points was that HP research would have to engage a lot more deeply in peer review through scientific and research journals. This idea makes so much sense- rather than a board at HP trying to work out whether an idea is really advancing the state of the art, the world will let HP know. This is open source/co-innovation thinking, and absolutely the right thinking to do.

I have regularly hammered IBM for having no “consumer” touch points. I came away from Bristol realising that HP has plenty, and its helping the company better understand Everything As A Service. HP owns snapfish. I knew nothing about this online photo storage app (being a dork I only know about Flickr, Smugmug and Picasa) - it has 50m users. So when HP researchers showed me an app that will examine your photo library, identify the faces in it, and visualise them accordingly, a light went off. I see- HP can actually use the stuff its labs are building. The same app could be bundled with a PC. I might not buy a machine because of it, but it would be a nice addition to the “service”.

I was most impressed though by HP’s sustainability guru Chandrakant Patel. The guy is awesome. I wrote his pitch up here.

HP has more skin in the game here than you might think - because of its printing business. While HP didn’t use the term Bit Miles it did talk a lot about “Long Tail Printing”. That is, digital printing at the point of use, avoiding the need to pulp a bunch of copies of some book or magazine noone ever read. Bear in mind that print technology is now moving into three dimensions, so you can potentially print objects not just characters on paper. The potential for print and micro-fabrication to reduce transportation cost is vast. Chandrakant talked about the need to create an “IT ecosystem” for the printing industry, to ensure it is carbon positive rather than negative. The HP Labs’ approach he said was to replace conventional supply chains with sustainable IT ecosystems.

Of course not everything in the vision is new. On the contrary:

“We need to leverage the past to create the future.”

One of the key problems with the 98% is the complexity of the metrics involved. How do we really know, asked Chandrakant, that the carbon used to create the Halo video conference wasn’t greater than the flight he chose not to take? There is a need for irrefutable metrics. And we don’t have 15 years. HP Labs is now working on prototypes to model and predict the impact of different re-engineering strategies, then measure and monitor the results. “These tools”, said Chandrakant’s UK equivalent Chris Preist, will analyse consumption of available energy and greenhouse gases across the lifecycle.

HP’s vision here is nothing less than to give businesses the tools they need to simulate the greenhouse impacts of potential new products and services. What if I used IBM tools here, or a BT network? What if I chose Apple hardware over Windows laptops? And so on.

Actually that took me ten minutes but at least its off my todo list! :-)

CloudCamp Comes to London:

cloudcamp london

Over breakfast on March 13th, we started to work out 15 Ways To Tell Its Not Cloud Computing. It was a good discussion, there is often plenty of truth in humour (just as there is often a kernel of truth in a rumour). One of my partners in crime that day, Alexis Richardson of Elastic Server has now gone about 500 better, and is bringing CloudCamp to London, in conjunction with a bunch of partners. The idea is to be as unconferency as possible. We’re all experts, or we’re all dunderheads. We’ll be using Open Space methodologies accordingly.

Basically we want to foster a conversation about Cloud Computing, how it can be applied, what the advantages are, who the best providers are, and what it means to your organisation. There will be lightening talks, but more importantly free beer and pizza.

There is talk of running the event outside, but it being London, you can be assured it will tip down if we do… It’s July 16th starting 1t 6:30pm. Please register here.

Silverlight 2.0, Deep Zoom, Client-side scripting and the Future of Archiving

In which we interview Brian Goldfarb. Its a pretty decent show. The discussion at the end gets interesting, when Bryan claims Ruby developers really don’t want to work with Javascript. In my experience most Rails devs are more than happy to see Javascript as part of their arsenal. But then they’re largely targeting browsers that actually support Javascript, unlike IE6, for example.

The Designer Who Gave Us Fail Whale and Showing The Whale

 

I had assumed Twitter was using designs it had paid for for its Fail page, but apparently not. Courtesy of John Wilson’s comment on my last blog about Twitter, a couple of hours ago, came news the famous whale is a stock shot from here.

Yiying Lu is the fantastic designer behind the image we all see, and love, so much. Frankly I think Twitter should use more of their work to give us some variety, and also give Yiying a lot more credit. How about a service sleeping owl?

With respect to celebrating downtime here is a link to the Fail Whale Fan Club (”Celebrating Twitter and our favorite error page cetacean“), funnily enough it seems it was started by someone I follow, Sean O’Steen. I love the fact Tom Limongello sent the Twitter team t-shirts!

He told them:

FailWhale is quickly becoming a brand, and that is a very good thing. Twitter has proven that it does a great job of bridging the online world to the physical world, in fact it’s better than any service I’ve ever used. I wore my FailWhale t-shirt at Internet Week in NYC and that simple test returned an immediately favorable response. I wanted to help the FailWhale succeed as a symbol, even though what it represents has not been completely defined (also a good thing, because you can define this to benefit Twitter’s image as a service that supports all of its users).

Perhaps one day we’ll be able to rename it the Scale Whale… but I am not going to hold my breath. Best of all- guess who is now on twitter… Yiying. Subscribed. I must admit I am thinking about Greenmonk logo possibilities… Certainly Yiying is now going to win a lot of work based on Twitter’s use of the image, which is GREAT, and proves the value of open content. There are so many great angles to this story. A community taking control of the brand, an artist creating a visual language that defines the way we talk about a service, the celebration of downtime that I mentioned earlier…

The Fail Whale is a classic social object. I wonder what Hugh MacLeod will make of it. 

I think a lot of web people are going to start using Yiying’s art. I know its now on the agenda as an image source for a chinposin Friday. To Whit To Who.

 

 

In Praise of… Downtime. Twitter as Phenomenon

 

Its not news that Twitter has problems with availability. What I think may be news however is the community’s reaction to it. Sure some people have been abrasive and abusive. Some of have claimed they’re going elsewhere (anyone for Jaiku or Plurk?) But they keep coming back. One thing we should all remember though- Twitter is still free (for now).

Me - I have to admit I kind of like the downtime. I don’t even mind when Twitter takes away the Replies tab, because the frankly rather spiffing summize search engine keeps me in the conversational loop. Sidenote to Jeff Bezos, who just invested in Twitter. Get the gang to buy summize as the start of a rollup: it is the killer app for Twitter. Its a reason to be on Twitter, rather than the other way around. If you’re in marketing go search for your product brand on there; its what Google Search used to be like before it became a good corporate citizen and started deprecating bloggers (its ironic that Goog became so much more like Yahoo just as it killed it).

But back to downtime. I now have a far more Zen-like appreciation of twitter as a medium.  If I miss a message so what. Its behind me. That is the way of flow. Rivers chatter, but you don’t have to listen to every eddy across every stone, every leaf in every whorl.

While some are freaking out about twitter others are becoming Stoic.

This week though I realised Twitter really has taken the appreciation of downtime to a new level. Not only do we appreciate it, we celebrate it. Exhibit A is the t-shirt above. I can’t think of any other service where downtime became something to enjoy, where it was so embedded in the platform’s success.

So Choose Life. Choose a platform. Choose Downtime.

This is a Web 2.0 pattern turned in on itself.

 

Zazzle calls its an Official Fail Whale t-shirt. Do they really have Twitter’s permission for the use of the image. Just wondering. 

Oh yeah one last thing I was supposed to mention in this post. According to summize @qrush was responsible was the first recorded use of Fail Whale.

Adobe RedMonk SAP Enterprisey Nanoconference: London July 11th

I am always interested in community spanning, and one of the most interesting community mashup dynamics of recent times has been SAP meets scripting language hackers, exemplified by this SAP T-Mobile case study I wrote up. So what does this have to do with Adobe? Well, Adobe Flex is making a nice home for itself in the SAP ecosystem, because its a solid and rapidly productive platform for application re-skinning, improving native SAP UIs and so user flow. Plenty of Java developers are also now learning Flex for similar reasons, and we’re beginning to see more corporate adoption.

Anyhow, on July 11th, James Ward IV, one of Adobe’s evangelists, is in town, so I thought why not bring some different folks together to discuss enterprisey opportunities for Adobe and so on. Given Jim can talk fluent Flex, but can also spell Spring and Hibernate, which are now deeply entrenched in UK financial services, I thought we could establish a pretty nice quorum.

The event will be Adobe’s new offices in Regent’s Park. There may be some coding, but mostly I just want to bring some smart people together, to learn a little. I am inviting some people, but if you’d like to come please do let me know. And yes I did realise that if the event is turned into an acronym its ARSEN. What is a nanoconference? No idea - I just figured it was even smaller than a micro-conference. This one is definitely an intimate event.

On SOA, Broccoli and Ice Cream

Broccoli and ice cream is a phrase Thomas Otter, enterprisey HR Maven, Gartner Analyst, cycling nutter and all round good egg, uses a lot in talking about enterprise software. He grew up with it -

“you can’t have your ice-cream without your broccoli.”

As systematic viewpoints translates it:

“broccoli - not as much fun as ice cream, but way more nutritious.”

For further broccoli and ice cream reference material see this podcast with SAP’s Jeremiah Stone. Anyway, I was re-reading Enterpriseyness the final sequel or my secret demo weapon exposed from Thomas recently (yes because someone tweeted it!), but unlike when I checked it our in 2006 I actually watched the video this time. Very funny, and to me it just shrieked SOA.

SOA should not just be about ice cream. It requires work, discipline and analysis. Its no good just leaving your existing business processes wheezing away, having a crafty cigarette behind a gleaming portal (looking a bit like Dennis in fact). That’s not SOA. SOA means diving into your broccoli, doing the work to really understand how things fit together, what can be jettisoned, what your data models are, how they can be usefully extended. Hey with SOA your enterprise architecture might even give up smoking.

Forget the ice cream. Real World SOA is all about the broccoli.

Bad Behavior has blocked 0 access attempts in the last 7 days.

Close
E-mail It