Lukas Svoboda: Strategy, Architecture, Governance & Technology

I must say, InfoQ is doing a really good job of covering enterprise development with plenty of interesting articles on SOA. There is a really good interview from Jim Webber from Thoughtworks (The same company Martin Fowler works for these days). He talks quite a bit about MEST (Message Exchange State Transfer) and an useful "standard" to support MEST called SSDL. The main jest of SSDL is that it is a replacement for WSDL (Web Services Description Languages) based on SOAP that gets you out of the Request/Response hell that WSDL creates. SSDL allows you define much more complex web service interactions such as sending two requests to a web service and get 5 responses back. It's also interesting to note that there is a SSDL open source implementation done by Patric Fornasier of National ICT Australia as a research project called SOYA. It's built on top of Windows Communication Foundation and looks very interesting. Suffice to say I'm going to be be playing a bit with this shortly.

The interview with Jim can be found here: http://www.infoq.com/interviews/jim-webber-qcon-london

More on SSDL can be found here: http://www.ssdl.org/

Lastly, you can find SOYA here: http://soya.sourceforge.net/

Interesting to see that Telelogic is offering a Carnegie Mellon certified Enterprise Architects program in this part of the world. The program consists of three courses (and I imagine some certification) to end up becomming a Carnegie Mellon certified Enterprise Architect.

Looks interesting, but certainly a big dent in any training budget.

Some cool comments I really appreciate from Mark Carroll on the recent ARCast I did recently with Ron Jacobs. Mark Carroll for those who don't know is the Architect in Microsoft’s Developer and Platform Strategy group here in New Zealand.

It has been a interesting week for articles on Computerworld I must say. It's been a while since I read it mainly because I've been very busy on a customer assignment and I was in recent times a little disappointed at the accuracy of some of the reporting. So I'm certainly finding some of the more controversial articles quite interesting this week.

Another one that stands out is this article on Poaching Talent, profiling comments on Simpl's CEO Bennett Medary's new Blog. The article quotes some interesting comments such as:

Medary says "vendors with the deepest pockets target and 'buy' specific talent required for immediate deployment to meet new or unfulfilled commitments. This often has the secondary effect of wounding the competition (and their customers) — killing two birds with one stone.

"Neither response is good for the industry or its credibility with customers!"

In a remark that appears to be targeted at Gen-i he comments: "Vendors with 50 - 60 vacancies each month do not have time for graduate programmes and other long term training/recruitment approaches! We have had a number of key staff specifically targeted by one large vendor recently, and my industry colleagues assure me that we are not alone."

Medary concludes offering to sign mutual non-poaching clauses with anyone "and publish who is and who is not willing to do the same".

Signing non-poaching clauses is no solution in my book. I think companies (and this is not targeted at Simpl) have to think seriously about how they look after and develop their staff. Medary starts to hit it on the nail with the comment "Vendors with 50 - 60 vacancies each month do not have time for graduate programmes and other long term training/recruitment approaches". I think this is exactly right - IT services companies must be putting in plenty of time and effort in keep staff by developing them. I think as long as someone is getting good money and good development then most people generally don't leave where they are working. Me personally I've found many of the large NZ SIs I've worked at are actually quite poor at growing peoples careers or they only do it with a small "key" subsection of their staff.

My experience as a manager in this industry is that if you don't provide challenge for good workers they will end up leaving for challenge somewhere else. When I ran the Microsoft practice at Optimation this was one of my primary concerns - get good work for people that’s a bit of out the box and make sure people have specific tailored career development programs for the individual. It's definitely no use taking a once size fits all approach. One tool I used quite a bit was learning profiling - i.e. working out the ways people learn knowledge. I figured that if I could work out how people learn, then I could probably work out a good development program. Some examples of this practically: One team member who learnt by working on the job doing super challenging customer work and had no patience for training and another one who liked solid work, but preferred structured training courses as he typically could apply better if he had the underlining theory drilled into him. As you can imagine I put together quite different development programs for each person.

A wise man in this industry once taught me that when people come work for you that if you develop them for their life careers then they probably will want to stay working with you for a good while. I truly believe this from firsthand experience and I honestly believe this is the best way to stop poaching. Oh and share options don't hurt either not that any System Integrators in NZ have caught onto this.

Knowing Rod, knowing what Xero is up to and knowing that SaaS is still proving to be a hard sell to many people, I find this Computerworld article very interesting where Rod disputes a US analyst’s take on SaaS. The analyst recently visited New Zealand.

Personally I generally agree with the Analyst's take on SaaS. It is still a low priority for larger firms in general and I still think SaaS is mainly stuck in the SME segment. Certainly my reading in New Zealand at the moment.

I personally think the SaaS movement will just end up forcing two things:

1. Dominant software sector incumbents to provide low cost/opex offerings (Watch the SaaS start-ups sweat when that happens)

2. More software gets offered as a web solution in addtion to desktop solutions. I think Google is probably the best model for this at the moment. They have really excelled in this area. It reminds me in the late nineties when Oracle was rabbiting on about thin clients and the end of Microsoft's dominance in software commodities. Well it still hasn't happened and Google still hasn't made it happen it despite producing some very good SaaS software. The interesting things to watch are the fact the Microsoft is investing big time in software frameworks that provide high useability on the internet. You can bet your bottom dollar they are hedging their bets and co-developing all the future products using this technology.

14-Nov: UPDATE FROM ROD (From Comments): "The article in computerworld did not accurately reflect the discussion. I agreed with Ken on many thing sand he agreed with the potential of SaaS especially in the SMB space".

Interesting comment from Rod there - Funny how computerworld distorted things! Rods comments sound much more like what I'd expect.

I've posted another architecture podcast (ARCast :)) on www.arcast.co.nz - this one is a recording I did with Ron Jacobs as part of his recent trip to New Zealand.  You can find the specific entry here or go straight to the podcast feed.

So as part of the Movember initiative for Men's Health I'm growing a wicked Mexican/Hulk Hogan Mo. It's all for a great cause and I will be posting my progress pictures on www.kiwibohemian.com.

Here is the official blurb - and if you can find it in your heart to sponsor me and donate a little money I'd be very appreciative J. No amount too small J :

Hi All,

During Movember (the month formerly known as November) I'll be growin a Mo. It's all for a good cause - men's health and the fight against prostate cancer. Why...

• Every year in New Zealand 2,656 men are diagnosed with prostate cancer and about 600 die of the disease, making prostate cancer the second largest cause of male cancer deaths, after lung cancer.
• The average life expectancy of a man in New Zealand is 4 years less than a woman.

To sponsor my Mo please go to http://www.movember.com/nz/donate, enter my registration number which is 128658 and your credit card details. Or you can sponsor me by cheque made payable to the "Prostate Cancer Foundation of New Zealand" clearly marking the donation as being for my Registration Number: 128658. Please mail cheques to: Movember, PO Box 87 150, Meadowbank 1742, Auckland. All donations over $5 are tax deductible.

All donations are made directly to the Prostate Cancer Foundation of New Zealand who will use the funds to create awareness, increase support networks for those men who suffer from prostate cancer, fund research and scholarship programs.

For those that have supported Movember in previous years you can be very proud of the impact it has had and can check out the detail at: Fundraising Outcomes.

Movember culminates at the end of the month at the Gala Partés. These glamorous and groomed events will see Tom Selleck and Borat look-a-likes battle it out for their chance to take home the prestigious Man of Movember title. If you would like to be part of this great night you'll need to purchase a Gala Parté ticket.

Cheers,

Lukas

More info is available at www.movember.com.

Movember is proudly grown by Holden, L&P, The Rock, Tui and Schick.

Movember a proud partner of the Prostate Cancer Foundation of New Zealand.

I've blogged a bit about using Wiki's recently and for my current customer I've been using Jive Software's Clearspace X (Version 1.6). This was certainly not a personal choice but rather a case of using an existing Wiki and social networking environment.

Over the last few weeks I've used Clearspace extensively to create an methods framework.  I would have created over 40 documents in the wiki and collaborated with several people on another 15 or so documents.

It's definitely been a mixed bag as I do like Clearspace and how it designed. I do like the simple workflow engine and the nice use of Ajax. I have to say though Clearspace has quite a few niggly bugs that make even basic content creation in Wiki documents difficult in certain circumstances and there are a couple of missing features which would really help me. I'm not sure how much of it is to do with my customers setup of Clearspace or direct issues in Clearspace itself.

Judge for yourself as I thought I would publish some my issues (and some workarounds) to hopefully help other people (Note these are very clearspace specific):

1. There are random problems with some of the text editors particularly for the homepage content - the general document wiki editors have less issues but I've seen some funny formatting problems with the formatted text box in the homepage of my wiki space. One strange thing I found is that once the formatted text box on the homepage got over about 6 paragraphs I could no longer edit it. It just would display the edit box but with no text to edit. I added three formatted text boxes to the homepage to get around this.

2. There does not seem to be any easy way to consolidate an entire wiki site into one document in Clearspace. My work around for this was to spider the site and run a PDF converter offline. Not ideal but it works.

3. In some cases images in pages don't resize properly and there is wrap around scrollbar that occurs. Workaround is to manually resize images or remove any URL stuffing in between the image formatting.

4. In general image resize after upload produces strangely grainy pictures. I know this could probably work a bit better but I've seen this on other wiki's too. Workaround is to create/resize your images to smaller sizes.

5. Wiki Images do not appear in PDFs. This is the number one issue for me - consistently PDFs from Wiki pages do not have images I upload to the wiki pages. It only works if I add images the first time I edit the document. Any images uploaded after the first edit while they appear in the wiki page do not appear in the PDFs. Very strange bug...

6. Spell Checker issues - I've stopped using the spell checker as it every time I use it ruins all my formatting and just rolls all the content together as one long string. Very weird, but this is a consistent problem. I couldn't seem to find a workaround for it.

7. Widgets errors on home page - widgets positioning getting mixed up. I've had this happen a few times when I setup a home page. Widgets would end up in place different to where I placed them. It seemed quite random and at other times it was fine.

8. Wiki pages do not like file URLs (e.g. file:// rather than http://) and once placed in and saved in a wiki page, the page just drops the URL. Very nasty. The workaround was to drop in an html page as a attachment. The html file when opened redirects to the file:// resource. Messy, but works.

I have used Microsoft Sharepoint 2007 and Confluence for creating similar content - It's my belief they are better products out of the box wiki wise. None the less, I have completed my framework using Clearspace and will continue to use it.