|
|
Posted by Dc on 04/06/05 17:05
David,
It sounds like you're really painting yourself into a corner...
Why don't you do one thing I suggested:
1) calculate the amount of time it would take to build all the form
widgets from scratch. Then put a dollar amount on it and put it next to
using the form widgets in Smarty.
It's a start.
DC
David Ngo had written:
> We are not talking about building a new template system here. What
> they are proposing is just using html/php as a template which will be an
> include file. We would also then have to come up with our own functions
> for displaying listboxes, radio buttons, tables etc. Yes that sounds a
> lot like Smarty, but the argument is that they would not have to learn
> Smarty that way and the php/html templates would be much simpler since
> everyone at work knows php but I'm the only one that knows Smarty. Our
> new programmer claims that she can write the xml caching in 6 lines of
> code using a harvester. I find that very hard to believe but maybe she
> can? I really cannot rule that out immediately without seeing the code,
> which by the way has not been written yet.
> Unfortunately I do not have any actual workplace examples where
> Smarty has saved me that I can share. I know that it makes my coding
> easier and I can pretty much cut and paste presentation logic at will
> but this is not the hard physical evidence my boss is requiring me to
> come up with . We have had problems a few times because the designer was
> designing header/navigation/footer templates in plain php/html and I had
> the same templates for myself in Smarty. This caused some navigation
> links and pictures to be inconsistent since some pages would call
> different templates. And the conclusion everyone came to is that Smarty
> was at fault. Though really this problem would occur with any site that
> has 2 different sets of templates, that kind of defeats the point of
> having templates.
>
> DC wrote:
>
>> David Ngo had written:
>>
>>> At my work I work with a html designer and one other programmer. Both
>>> of them refuse to learn to use Smarty and keep arguing that we should
>>> do away with it.
>>
>>
>>
>> Don't point out to them that they are arguing from ignorance. That will
>> make it hard to sway them to your point of view!
>>
>>> My boss's take on it is that Smarty adds complexity with
>>> little benefit.
>>
>>
>>
>> Um, having a team of programmers developing and testing a solid code
>> base is a pretty big benefit. Who are you going to get to test and debug
>> the proposed XML caching system? How much time and money will it take to
>> build? Is it a critical part? If so, why not use a mature
>> compiling/caching system instead of a new caching system?
>>
>>> Their solution is to just do away with the templates and
>>> instead have php include files with html in them that will act as the
>>> templates.
>>
>>
>>
>> That sounds a lot like Smarty to me! "Php include files with html in
>> them" is just another term for template. Besides, you can avoid using
>> the complicated stuff pretty easily. And then it boils down to building
>> html files with smarts.
>>
>>> We also have a proposed solution using xml that can replace
>>> the caching options that Smarty has built in.
>>
>>
>>
>> Yes, the key phrase here is "proposed." Why use vaporware when you can
>> use real, battle-tested code?
>>
>>> I have to come up with a
>>> few examples where using Smarty is a better solution.
>>
>>
>>
>> Why don't you share them here?
>>
>>> Does anyone have
>>> any extremely convincing examples I can use to change these people's
>>> minds? All of the examples I have seen on the Smarty site and through
>>> google are very simplistic examples that could easily be replaced by
>>> php includes.
>>
>>
>>
>> I think the key is to find the breaking point for the application you
>> are building. That is, where has Smarty saved you in the past? And what
>> capabilities are you going to want to add to your custom solution that
>> Smarty already has.
>>
>> Another important point that people often overlook is that Smarty is
>> very widely known and used. That will be a critical selling point for
>> including it in your system. This is what I call "universality." It's
>> the likelihood that others will know Smarty when it comes time for you
>> and your co-workers to move on. If you don't go with Smarty, the company
>> is left with a highly specific, customized solution with (very likely)
>> no documentation, and that no one know how to maintain or develop.
>>
>>> I do not want to move away from Smarty because I feel it makes
>>> my code better structured, easy to read, and just forces me to have
>>> better coding habits.
>>
>>
>>
>> Those are great reasons! Now translate them into dollar amounts for your
>> boss. Do a conservative estimate of the time it would take to replicate
>> some of the Smarty technology in your custom system, then multiply that
>> by your estimated hourly wage. Then multiply that by 2-4 because we are
>> all optimists and underestimate the time it takes to do anything custom
>> built!
>>
>>> However how do I explain to someone that has never
>>> used Smarty and refuses to learn it that it is the way to go?
>>
>>
>>
>> Let them drive it. You can't argue against someone who has a closed mind
>> and no direct experience.
>>
>> Come up with an ordered list of criteria for evaluation: say, initial
>> development time, application maintenance and expandability, code speed,
>> separation, learning curve, documentation, "universality", etc... Each
>> person will have different ideas of what should be at the top of this
>> list but will basically come down to "company makes more money" and
>> "workers suffer less pain." But, do the ranking separately for each
>> stakeholder, and don't share rankings until the test is over.
>>
>> Then, clear some time in your dev schedule and devise a test
>> mini-project to develop in Smarty and your custom template environment
>> in parallel. This should only take 1 day to build both. Then look at
>> both solutions with your criteria in hand. Whichever satisfies the
>> critical criteria the best (across all interested parties) is your
>> winner.
>>
>> The key is to come up with what is important to you and your company and
>> rank it and then take a realistic view of the results.
>>
>>> I fear
>>> that if I can't come up with some good examples we will switch to the
>>> system of using php includes as templates which I do not like at all.
>>>
>>
>> HTH,
>> dan
>>
>>
>
>
Navigation:
[Reply to this message]
|