Comments for About us, clouds and software

Comment on Is multi-core programming still hot? by Guancheng Chen

Guancheng Chen — Sat, 28 Apr 2012 15:34:27 +0000

Hi Andras, I totally agree with you. When I started to focus on Multicore Programming, everyone said every programmer need to know it. However, currently, multicore programming is still not in the mainstream. Think out of the box is very important in this case, since the thing that you believe is important may in fact be irrelevant (yes, if you have worked on Computer Architecture for a long time, then Multicore is definitely a big thing). To really build a success business on new technology, we really need to think about what are really needed by the customers.

Comment on The Cloud in 2012 by Diane Mueller

Diane Mueller — Wed, 04 Jan 2012 00:11:53 +0000

Nice work on AWS cost analysis – I agree with your assessment that we will be seeing a slowing of the public IaaS, and seeing more growth in the private cloud space – as this is what we’ve been experiencing already towards the end of 2011. Bare bones is definitely not enough, and security issues will continue to drive adoption of private clouds along with value-added layers such as PaaS and other services. PaaS’s ability to deliver highly curated multi-tenancy containers to the Enterprise enabling much more automation and flexiblity than bare bones AWS instances or vms – is going to be a driving force for private cloud adoption in 2012.

Comment on Old story, told again (with a twist) by Treadmill reviews

Treadmill reviews — Fri, 02 Dec 2011 09:28:49 +0000

I suggest adding a “google+” button for the blog!

Comment on Monetizing social networks by Elric

Elric — Fri, 30 Sep 2011 01:27:03 +0000

You’re the one with the brnias here. I’m watching for your posts.

Comment on Is IaaS fading away? by Lachlan

Lachlan — Thu, 29 Sep 2011 17:27:36 +0000

Interesting. Currently I’m going for BSIT in Networking. Cloud computing is really big inovation and yet so many companies can not understand how easy it will make their jobs.

Comment on Monetizing social networks by alper

alper — Fri, 16 Sep 2011 07:12:49 +0000

Hi Andras,

I listened your presentation at SICS’s event at Kista this week. Could you please mail me your presentation slides, it was a great work.

Alper
alpercelk@gmail.com

Comment on Monetizing social networks by Olin Neumann

Olin Neumann — Thu, 08 Sep 2011 14:05:38 +0000

I wanted to let you know you wrote a great article.

Comment on Is the future BrowserOS? by Chromebook, anyone? « About us and parallel programing

Chromebook, anyone? « About us and parallel programing — Thu, 12 May 2011 11:18:27 +0000

[...] Google would obviously not agree with the conclusions of my post a couple of days back – they’ve just launched yesterday their ChromeOS and a couple of [...]

Comment on How the new conductor wants to make the elephant dance again by Mike

Mike — Thu, 10 Feb 2011 21:56:54 +0000

Your definitely not alone in the way you think.

http://justredoit.wordpress.com/2011/02/10/what-will-nokia-do/

Comment on How the new conductor wants to make the elephant dance again by Benny

Benny — Fri, 04 Feb 2011 19:35:24 +0000

I just started writing some Meego app. The environment looks great. The software is available, great SDK.

Hence, the strategy is easy: Meego as the new all Nokia platform, Symbian being phased out, and an allience with Microsoft to run Qt apps on WP7 phones. Silverlight is not the future.

Comment on What you can learn by writing a book by Vajdi

Vajdi — Thu, 03 Feb 2011 06:55:58 +0000

Thank you!
It’s “Programming Many-core Chips”; it was commissioned by Springer. It should show up in on-line stores for pre-order in the coming weeks (due to appear in Q2)

Comment on What you can learn by writing a book by Softtalkblog

Softtalkblog — Tue, 01 Feb 2011 10:39:06 +0000

Congratulations! Writing a book can be an exhausting but ultimately rewarding experience. What’s the title and who is the publisher of your book?

Comment on A tale of the textile and software industry – part II by Vajdi

Vajdi — Thu, 09 Dec 2010 09:41:56 +0000

Your line of argument only holds if the condition is true on long term (5+ years). But, if the cost of relocation is compensated for by savings within, say, 5 years and the cost difference is maintained afterwards, it’s still a good business case. A global company won’t be planning with just 1-2 years in mind….

Comment on A tale of the textile and software industry – part II by Michael Williams

Michael Williams — Fri, 03 Dec 2010 11:45:58 +0000

Of course a global company may locate major development, manufacturing or even services to locations where they are cheapest. However, at least in the medium term, relocation comes with a cost, i.e. administration, training, organisation. If

Cost_of=relocation + Cost_of_doing_it_in_cheaper country > Cost_of_doing_it_at_home

then relocation won’t happen. I believe the cost of relocation is significantly higher than people first believe. But nevertheless we need to reduce the cost of doing it at home if we want to keep work in our western countries.

Comment on 1+1 of modeling by Vajdi

Vajdi — Tue, 09 Nov 2010 17:20:11 +0000

Well, requirement to model transformation is still far from reality (and it will take a looong time to materialize). As for testing, you can’t use the same model for code and test cases, as that inherently means you cannot detect faults. You need a decoupled model for testing.
Coupling design and implementation is real and that’s what I highlighted when saying that you can visualize the system at different levels of abstraction.

Comment on 1+1 of modeling by Mats Brorsson

Mats Brorsson — Tue, 09 Nov 2010 09:01:45 +0000

Aren’t modelling supposed to help out in bridging the gaps between requirement specification, design, implementation, test and maintenance so that you actually know that you are working on the same system in all these phases? A graphical interface can be used without having a model abstraction behind it.

Comment on A tale of the textile and software industry by Michael Williams

Michael Williams — Tue, 26 Oct 2010 10:30:52 +0000

The only way to keep software development in Sweden, or for that matter any western country, is that the cost of software development has been rationalised to the extent that it is no longer the major cost of product development. Outsourcing to “low cost” countries will always have extra cost associated with it. The lack of interactivity with developers has a cost, time taken for bug fixing (extra communication) has a cost, lack of flexibility has a cost.

If we can make SW development cheap enough, the additional cost of outsourcing will hinder outsourcing. I think we can do this, they key being the ability to program at higher levels of abstraction (functional languages etc). Of course this will have the effect that we will need far fewer programmers in the western world, but this is better than losing the whole of the SW development industry.

Comparision with the textile industry is wrong, there was basically no way to make the production of cloth and clothes cheaper except by outsourcing. I think we are not in the same situation as regards software.

Comment on How to make the elephant dance again by Johan

Johan — Fri, 10 Sep 2010 19:12:38 +0000

At least, now you can say – “I told you so”
http://tech.fortune.cnn.com/2010/09/10/can-stephen-elop-fix-nokia/

Cheers,
Johan

Comment on Net neutrality: it’s about scarce resources, stupid! by Johan

Johan — Fri, 10 Sep 2010 19:08:18 +0000

Andras,

Interesting read, but I could not disagree more. One could (and should) argue that the only ones that do get paid today is the operators – not the ones who provide the content that makes the networks attractive. The only reason people are willing to subscribe to the networks (and this is much more expensive in the US compared to Finland/Sweden) is the availability of mostly free newspapers, movies, music. The operators are already cannibalizing on the content providers and it would make more sense for them to pay back, not the opposite.

Cheers,
Johan

Comment on The world of computing, AD 2020 by Jakob Engblom

Jakob Engblom — Tue, 31 Aug 2010 18:12:36 +0000

Hm. Interesting idea. I am not totally sold on the idea of lots of hardware core/hardware threads scouting ahead for a main thread as opposed to a single powerful traditional execution unit. At least for things which are very branch- and icache-intense like virtual machines and virtual platforms, I would really like to know what would work and what would not.