I literally check developers code a lot when they first start. (2-4 weeks worth)
Make sure all the standards are being followed, then I let them go, and developers check each others code once they are in the company standards set. I find, that if you are introducing new standards to a company that doesn't already have them or has bad ones, get the developers to assist with the setting of the standards, that way they feel part of it, not something that's forced upon them. I fanatical about tabbing, but it's all in the greater good. Regards Dale Fraser http://learncf.com -----Original Message----- From: William Seiter [mailto:[EMAIL PROTECTED] Sent: Monday, 8 October 2007 12:59 PM To: CF-Talk Subject: RE: CF Coding Standards You are the lead, you see a problem with not only the current code but future code put together. You are introducing standards to the group. This is not at all a bad thing. You will probably be heralded in the future for your attempts now, however despised you might be for it now. I would suggest doing 'spot checks' on other people's code, or 'peer reviews' on their code so that you can verify that the standards are being kept to. You might even do this as a 'group' event. Where the group of developers gets together and puts 1 programmer on the 'hot' seat. The programmer will 'run through' his/her code and explain what it is doing and why it is there. You could critique the code based on ample commenting and standards adherence, and you can have the group come up with plausible solutions for 'heavy' code or code that is 'muddled'. This would be a great exercise to help the 'junior' coders get better as well as the 'senior' coders to hone their skills and 'show off' a little. To get everyone off on the right foot you might even make the first few 'sessions' regarding a custom tag or UDF that you can download from the exchange. Send it out to everyone and have everyone 'change it' based on their programming skills but to your new standards. Then you can run through the original example and open it up for discussion. Just an idea... Can you tell I am 'dying' to get out of my '1 guy' shop and join a bigger team? William -- William E. Seiter Have you ever read a book that changed your life? Go to: www.winninginthemargins.com Enter passkey: goldengrove -----Original Message----- From: Daniel Kim [mailto:[EMAIL PROTECTED] Sent: Sunday, October 07, 2007 3:07 PM To: CF-Talk Subject: Re: CF Coding Standards I recently started a position as a technical lead at a new company. Most of the development is done on a few intranet/extranet sites, so the code (much of it spaghetti) is touched by a lot of different developers. I've recently set up a wiki that contains coding standards, and used a lot of the LiveDocs recommendations (with a "References" section indicating sources). I've gone the route of placing these standards under some general categories, such as "Security", "Performance", and "Stylistic". The "Stylistic" category defines things such as capitalization and indentation. It may have been a bad idea, but with this category, we came to a lot of these standards based on the consensus of the group - most of whom maintain the coding style of the CFWACK, e.g., title casing function names, uppercasing CF operators, etc. Wondering if any of you would think it is beneficial (or too draconian) to have these "stylistic" standards in place, even if it isn't an individual's preference, to promote readability, consistency and prevent constant reformatting (annoying when doing diffs), especially in an environment where different developers are touching a lot of the same code? On 10/7/07, ColdFusion <[EMAIL PROTECTED]> wrote: > > Thanks, I will make note of that. > > -----Original Message----- > From: Adrian Moreno [mailto:[EMAIL PROTECTED] > Sent: Sunday, October 07, 2007 2:28 AM > To: CF-Talk > Subject: Re: CF Coding Standards > > >I recall a benchmark results from along time ago indicating that if > >your doing more than 3 CFIF/CFELSEIF statements it was better for > >performance to use CFSWITCH. That was discussed here on the mailing list > of > HoF. > > > http://webapper.net/index.cfm?fuseaction=Fuseblog.ShowComments&ArticleID=200 > 60727042244 > > Comparing cfswitch/cfcase vs. cfif/cfelse: They found that when comparing > string values, if/else is much faster then switch/case under load. When > comparing numeric values, switch/case is faster. > > -- Adrian > > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Download the latest ColdFusion 8 utilities including Report Builder, plug-ins for Eclipse and Dreamweaver updates. http;//www.adobe.com/cfusion/entitlement/index.cfm?e=labs%5adobecf8%5Fbeta Archive: http://www.houseoffusion.com/groups/CF-Talk/message.cfm/messageid:290506 Subscription: http://www.houseoffusion.com/groups/CF-Talk/subscribe.cfm Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
