dougherty distilled - All Comments

re: A Scrum-ptious Methodology?

Rohan Ghatpande — Wed, 16 Jul 2008 05:12:46 GMT

Bryan,

I think this is a very objective evaluation of Scrum.

At Motorola we follow scrum for all our in-house development and since I am right in the thick of it and hence I can relate to your arcticle quite well. Here is how we tackle some problems you have pointed out.

Stand up meetings:

We have an estimates document which gets updated daily by the scrum master based of the hours quote given by the developer for the remaining work. Although I see your point when you say "Autonomy of the developers is a tenet of Scrum, yet developers have to check in every day. Or maybe you don't trust your team? That's one way it could look to a developer" but I can vouch for myself atleast that it has helped to keep the team focused on the Task at hand...I guess its just worked out for us. We have a team size of 6 including a Program Manager, Tester, Technical manager and 3 developers. We never had issues with members comming late as we use Netmeetings and conference bridges to conduct the scrum so in worst case a member can attend the call even if he is in a train comming t work.

Requirements aren't always perfect:

This is very true in our case most f the time we get request like "We want a better view of Sales". We tackle this in two tiers in between the user and developer. The Program Manager probes and suggests the user and refines the requirement to "Show us sales by States" and generates a document called as CRS(Customer Requirement Spec). This CRS is evaluated by our Technical manager (who happens to be our Data modeler) generates a SRS(Software Req Spec) saying 'use Tables A,B and C to ceate a DataView X to show sales by state'. The SRS is the starting point for the Developer which is quite clear.

I agree that there is a potential for miscommunication between CRS and SRS and yes it has happened before but fornately we have yet to encounter a major problem with it.

Long Term Planning:

This is one place atleast I feel that our process doesnt work so well. Although we try to divide and push the work in sprints I dont think it works in situations where we are rewritting a section in a better technology. I am currently being asked to rewrite some of ASP pages to ASP.net and due to the short 'sprints' there is hardly any time or chance to bring in the power object oriented framework that ASP.net provides.

So in essence Scrum makes me focus on getting the work done but not getting it done the right way. Sure it can be argued that I can give higher estimates but we all know planning and design phase has certain trial and error involved in it and i feel the pressure of daily updates fails to provide an environment for a developer where he can get to his creative side to come up with a state of art design.

These were some of my thougts based on your article. It would be nice to hear more thougts on possibly what in your opinioun we have got right or wrong so that we may evolve accordingly.

Feel free to get in touch with me your comments at rohanghatpande@gmail.com

Web 2.0 - Social Media - Internet News - Blogging » A Scrum-ptious Methodology?

Web 2.0 - Social Media - Internet News - Blogging » A Scrum-ptious Methodology? — Mon, 19 Nov 2007 00:29:59 GMT

PingBack from http://hyiplive.org/a-scrum-ptious-methodology

re: Custom Serialization and the Misbehaving Dictionary

Mike Gavaghan — Sat, 21 Jul 2007 03:12:04 GMT

Hmmm. Link got botched. Let me try it without the markup:
http://www.gavaghan.org/blog/2007/07/17/fixing-bindinglist-deserialization/

re: Custom Serialization and the Misbehaving Dictionary

Mike Gavaghan — Sat, 21 Jul 2007 03:10:09 GMT

BindingList is broken. I ran into a related problem with C# trying to couple BindingList with INotifyPropertyChanged. I'd get the ListChanged/ItemChanged events at first, but after serializing/deserializing the list, the events wouldn't fire. It's completely related to .NET failure to rewire the internal event listeners after deserialization. That's disappointing. This was my workaround: <a href="http://www.gavaghan.org/blog/2007/07/17/fixing-bindinglist-deserialization/">Fixing C# BindingList Deserialization</a>.

Thanks for the post.

re: Pac-Man for Mobile 5.0

david — Wed, 18 Jul 2007 18:35:39 GMT

hello, very good his articles about Windows mobile, I have some experiences developing games for celualr phones on java with j2me, but this of .net is new for my, i like to know if this arranged to help me in any problem that can present to me . thanks

re: New ASP.NET Dynamic Data Controls : Nifty, But Not Real World

Jay Basch — Mon, 07 May 2007 15:42:29 GMT

I found your review interesting and am glad you commented on this particular MIX session. I was looking at this session as more or less a "Proof of Concept" than an answer-all. From an application developer perspective I see this as unique set of tools for RAD techniques. The current feature set of the dataset model is to burdensom for the developer in its current state. I must agree with you that code generators do not always save time, but in this respect I believe this presents an aspect of descoverability that will aid application development greatly. I am going to keep in tune to the developments of this technology on my radar to see what happens. Once this and LINQ mature I can see major benefits.

re: New ASP.NET Dynamic Data Controls : Nifty, But Not Real World

Kevin Mao — Sat, 05 May 2007 02:31:35 GMT

Good Point that Dynamic Data Controls is not for Enterprise Application and there is no control on security and business rules to fill the enterprise requirements. Very easy example if I need to export the gridview to excel, how can I can with Dynamic Data Controls?

re: A Simple Configuration Pattern

bdougherty — Thu, 25 Jan 2007 14:09:21 GMT

Glad it helped. I'll try to revisit this and see what the deal is. Might be an even better way to clean this up.

re: A Simple Configuration Pattern

Cormac — Thu, 21 Dec 2006 23:06:43 GMT

Hi Bryan, a neat way of loading XML configuration and getting .NET to do all the donkey work. The only thing I noticed was that I had to use section.InnerXml, i.e.

XmlTextReader xmlReader = new XmlTextReader(section.InnerXml.Trim(), section.NodeType, null);

as the OuterXML was just whitespace. Everything else worked like a charm.

Thanks!
C

Q Thoughts

dougherty distilled — Thu, 08 Jun 2006 14:47:07 GMT

There's been a rumble on the street for quite a while regarding Motorola's new Smartphone, the Q.&nbsp;...

re: Upgrading to Windows Mobile 5.0 Smartphone

bdougherty — Wed, 31 May 2006 14:51:02 GMT

What I was saying is that the mpx220 comes with Windows Mobile 2003 OS which only supports Compact Framework version 1.0. To install the 2.0 framework, the Windows Mobile 5.0 OS would need to be there. Maybe technically the mpx220 can run Windows Mobile 5.0 as it's OS. I doubt it, but I don't know. Probably a Motorola question.

-Bryan

re: Pac-Man for Mobile 5.0

bdougherty — Wed, 31 May 2006 14:38:01 GMT

Hi, Jen. Thanks for checking out the article.

The server where the code downloads live was down for a couple days. The links should work now.

-Bryan

re: Upgrading to Windows Mobile 5.0 Smartphone

caleb — Wed, 31 May 2006 00:17:12 GMT

according to the link given, why can't the mpx220 operate on windows 5.0? isn't the link saying that for version 1 and 2, the smartphone can support 5.0 platforms? thanks

re: Pac-Man for Mobile 5.0

Jen — Wed, 24 May 2006 09:34:50 GMT

Hi,

I just tried to download the source code and the links don't seem to be working :-( Any suggestions? Btw- great article on coding4fun!

Thanks!

re: Upgrading to Windows Mobile 5.0 Smartphone

bdougherty — Tue, 18 Apr 2006 13:55:03 GMT

Unfortunately (I say this because this is my current phone, as well) the MPX220 doesn't support CF2 because it doesn't run Windows Mobile 5.0. Check out this link to see the full list of supported combinations.

http://msdn2.microsoft.com/en-us/library/ms172550(VS.80).aspx

-Bryan