SitePoint Sponsor

User Tag List

Page 1 of 2 12 LastLast
Results 1 to 25 of 42
  1. #1
    SitePoint Zealot
    Join Date
    Oct 2002
    Posts
    158
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    It seems as though everyone has their own framework

    .... it really does. There are an innumerable amount of people saying that they are working on their own framework.

    Is this a good or bad thing? It kind of sounds like it will make things very difficult when different apps/websites are rolled out using these different custom frameworks... but then again I guess I could be wrong since all frameworks will always incorporate the same things (only approached differently).

    This also makes me wonder if .NET framework has an advantage (market-wise) since it is a standardized and widely accepted framework.

  2. #2
    SitePoint Enthusiast
    Join Date
    Jan 2003
    Posts
    60
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    i noticed too

    I noticed this as well and here is what I concluded.

    There are WAY too many Content Management Systems, all of which implement their own API. This just shows the adolecence of the php community. It has a long way to go before it reaches maturity.

    Right now I am writing a new framework. You might say, great, another one. But no, I am modeling mine off of Struts, as close as I can get without trying to make PHP something it isn't (alot of the java code I am just tossing to the side since it deals with type conversion and all this business that php doesn't have to get involved with). Yes I know about phrame and yes it is pretty good, but I am hoping to go to the next level.

    What I believe is that everyone should be able to make their own web applications, but we should all be working off of a fairly tight set of frameworks, similar to jakarta. Most of the frameworks that are written suck and were a waste of time.

    While Struts is not absolutely flawless, it certainly has stood the test of time in very high pressure development enviroments and has proven itself as a model which works. There are better ones out there, so port those, learn from intelligent people instead of thinking you can come up with something better. If you are lucky, you just might change the way Java developers make the frameworks if you get involved. Just because they are writing in one language doesn't mean that it does not apply to php as well.

    I welcome developers to join me. I will have a sourceforge project up sometime soon, and it will be called (tenatively) Studs.

    Let's try to work together here instead of all going off into a corner and piecing together are own APIs. I also feel like we should build on the Eclipse library as a "common" framework, must like Jakarta's "common" distro.

  3. #3
    Currently Occupied; Till Sunda Andrew-J2000's Avatar
    Join Date
    Aug 2001
    Location
    London
    Posts
    2,475
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: It seems as though everyone has their own framework

    Originally posted by shoebox
    .... it really does. There are an innumerable amount of people saying that they are working on their own framework.

    Is this a good or bad thing? It kind of sounds like it will make things very difficult when different apps/websites are rolled out using these different custom frameworks... but then again I guess I could be wrong since all frameworks will always incorporate the same things (only approached differently).

    This also makes me wonder if .NET framework has an advantage (market-wise) since it is a standardized and widely accepted framework.
    This is a very true statement, which is why I have been crossing over from php to dot net...

    Although i'm another one of these people currently in the mix of developing my own framework before I move completely onto dot net. The reason being is that I think it is far more productive than PHP. Although PHP has a vast number of functions, I think quite a few could be collaborated to create some form of unified model, such as authentication.

    Please let us know when you intend on starting this project btw...
    Last edited by Andrew-J2000; Feb 22, 2003 at 04:16.

  4. #4
    SitePoint Zealot
    Join Date
    Oct 2002
    Posts
    158
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I have dabbled in .NET myself, but I really dislike IIS with a passion, so I quickly veered off and started doing OOP in php.

  5. #5
    No. Phil.Roberts's Avatar
    Join Date
    May 2001
    Location
    Nottingham, UK
    Posts
    1,142
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Im learning C#, but not for web applications....

    I actually use the base code from Invision Board as my chosen "framework". Its not really a proper framework and it wasnt designed for what I do with it. But I'm used to working with it and it lets me get stuff done quicker. Which is what matters in the end.
    THE INSTRUCTIONS BELOW ARE OLD AND MAY BE INACCURATE.
    THIS INSTALL METHOD IS NOT RECOMMENDED, IT MAY RUN
    OVER YOUR DOG. <-- MediaWiki installation guide

  6. #6
    SitePoint Wizard gold trophysilver trophy
    Join Date
    Nov 2000
    Location
    Switzerland
    Posts
    2,479
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Think there's two things in play here: first there's a library for solving all the common problems you encounter then there's a framework which is an "environment" that executes your code.

    The library in PHP is still work in progress. Overall there's a ton of free classes out there in PHP but (at a a guess) only 1 in 10 is designed well enough to be truly reusable without placing requirements on the code it uses. .NET has some value over PHP as it's got a well designed library, albeit missing some things like ftp and flash support. Personally though, think with PHP 5 this may well become an irrelevance as support for integration with Java, .NET etc will make it possible to use whatever library you choose.

    On the framework side, everyone has written a framework of some form, perhaps just;

    PHP Code:
    switch ($_GET['view']) {
        case 
    'home':
            include 
    'home.php';
        break;
        case 
    'news':
            include 
    'news.php';
        break;
        
    // etc

    The issue is having a framework that allows you to deal with many problems and allows you to code the way you want, and encourages efficiency.

    Neither PHP or .NET have solved this problem on the web: perhaps no one has. If you at the sample app for "how you should do it with ASP.NET" you'll see something that won't be winning any awards for design. IMO that structure is equivalent to the sort of code most PHP developers produce when they first get into templates, with a bit more OO thrown in to give a feel good factor.

    From what I've seen, struts is a big step in the right direction but I really don't anyone has hit on the perfect framework for the web yet (except possibly IBM with Websphere but I don't know enough detail).

    Being practical for a moment though, what we're really talking about here is an environment which makes it easy to build web applications; saves you time and takes away the "risk of failure" plus has a commercial aspect (backed by a company) so that it won't fall into the SF graveyard. For me that framework already exists in PHP eZ publish.

  7. #7
    Non-Member
    Join Date
    Jan 2003
    Posts
    5,748
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    mojavelinux - I would have to disagree with your point in everyone using the one 'universal' framework.

    If everyone had to use the same framework this would inhibit the learning process - it wouldn't encourage people to sit down and design/develop their own environment - the basic reasoning behind developing and programming would be lost.

    Also, you cannot expect one solution to solve ALL problems - there will be times where you would then need to use a seperate course to find another solution.

    I don't know about struts (Java ?) though I suppose I will at a later date if I get involved in Java to that level. I also agree with HarryF's comments about PHP5 which would take PHP to the enterprise level.

    This I cannot wait for, the sooner I can get my hands on this version, the better.

    As for .NET I do not really want anything to do with that - though it is a good thing to have PHP and Java working together finally.

  8. #8
    Mlle. Ledoyen silver trophy seanf's Avatar
    Join Date
    Jan 2001
    Location
    UK
    Posts
    7,168
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Originally posted by Dr Livingston
    This I cannot wait for, the sooner I can get my hands on this version, the better
    As long as you're willing to cope with bugs you can always play with the latest versions

    http://snaps.php.net/
    http://cvs.php.net

    Sean
    Harry Potter

    -- You lived inside my world so softly
    -- Protected only by the kindness of your nature

  9. #9
    SitePoint Zealot
    Join Date
    Aug 2002
    Posts
    178
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Originally posted by Dr Livingston
    I also agree with HarryF's comments about PHP5 which would take PHP to the enterprise level.

    This I cannot wait for, the sooner I can get my hands on this version, the better.
    http://snaps.php.net should provide you with a snapshot

  10. #10
    Non-Member
    Join Date
    Jan 2003
    Posts
    5,748
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Thanks folks, I am aware there would proberly be some bugs , thus I am waiting for the official release - sometime this spring I think - March, maybe April.

  11. #11
    SitePoint Enthusiast
    Join Date
    Jan 2003
    Posts
    60
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    not just one

    Dr Livingston, I didn't mean just one. It is perfectly clear that one framework for all won't work. What I said was "a close nit group" of frameworks, like jakarta. There are many jakarta frameworks, which includes Struts. They all build off of a common library, which includes BeanUtils, the Java core and javax, among others.

    Vincent has also designed a framework called Dawn, which he is considering for release, but has not yet decided on the license. He too implements a fairly similar MVC pattern, although one with a bit more purity. Point is, he built it off the Eclipse library as am I. Many of the ideas will be the same but perhaps only differ in the way they are carried out, such as the "View" part or the the "Model" part.

    Yes, it is possible to have a libarary which is reusable and no PEAR is not the answer. PEAR has no Iterator class, no Stack class, no "simple" HttpRequest, HttpResponse or Session class (yes it has flavors of them, but they do more than they should).

    I would never use asp.NET because Microsoft doesn't exist to me. But that is a personal opinion and by no means do I have to impose it on anyone else. By the way, do you have to pay for .NET and IIS webserver? Just curious.

  12. #12
    Non-Member
    Join Date
    Jan 2003
    Posts
    5,748
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    About IIS you can get version 5 with Windows 2000 Pro, you need to buy .NET don't you ? I'm not sure...

    ...as for Microsoft and .NET; I don't like the idea either and I am fully rooting for Java myself - all MS software in the past has had security issue's of some sort, and .NET is no different.

    About your idea on frameworks - I missed the point.

  13. #13
    "Of" != "Have" bronze trophy Jeff Lange's Avatar
    Join Date
    Jan 2003
    Location
    Calgary, Canada
    Posts
    2,063
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Windows XP Pro also has IIS5.1, and .NET has 6.0
    Who walks the stairs without a care
    It shoots so high in the sky.
    Bounce up and down just like a clown.
    Everyone knows its Slinky.

  14. #14
    SitePoint Enthusiast
    Join Date
    Jan 2003
    Posts
    60
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    so why...

    So why would people want to give money to that *$%# company rather than use what is out there, open source and with security held high up in priority?

    I am not saying Java is the answer, or else I would be actually use Tomcat and I would leave this forum. Java is just too awkward of a language for me to make websites in (type conversion and many library dependencies to use core functions) but I do believe the ideas are strong. I don't think ASP.NET has anything that jakarta hasn't already hit. I do believe however the PHP projects fall well short of what both ASP.NET and jakarta offer and that is why I am encouraging a community to be established where we can but these ideas (such as MVC) together into a project. HarryF does a great job with the website phppatterns.com and it would be a constant reference. But it just seems like a lot of people are going in a lot of different directions.

    Using a framework such as Struts, phrame or my new one Studs (all basically the same) you can use XSLT to format XML data or you can just use PHP templates. Point is, a framework is much more broad than these specifics. A framework is how you think about the code. Perhaps the PHP community has the word "framework" all backwards when we consider the name of a project.

  15. #15
    Non-Member
    Join Date
    Jan 2003
    Posts
    5,748
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Ummm... never thought about it that way. Personally, I am only interested in developing a method whereby I can easilly create and maintain a dynamic web page(s) using PHP.

    At the moment, I see using XML/XSLT as a way of doing this, in more ways than one. The bonus being, using XSLT I can transform a basic XML template to another format other than (x)HTML.

    If at some point in the not too distant future, I can implement Java then all the better. At the moment, PHP may not be on the same level as Jakarta/ASP .NET but we are all hedging our bets on PHP5 and the new Zend Engine 2.0 changing all that.

    PHP is open source and a popular language and things can only get better...

  16. #16
    ********* wombat firepages's Avatar
    Join Date
    Jul 2000
    Location
    Perth Australia
    Posts
    1,717
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    But it just seems like a lot of people are going in a lot of different directions.
    aye , and that (for me at least) is the point, or more to the point , not the point

    PHP is not JAVA , nor is it .NET (nor indeed is .NET yet but anyway) , aside from its ease of use PHP's no 1 'selling point' has been its flexibility, no constraints, i.e. if you want to clone struts etc , you can , or ,you can build up a collection of code for re-use and rewrite your framework to suit the application at hand.

    The complexity of such frameworks are both their pro's and their cons , in a similar fashion that some templating systems become so complex that perhaps at the end of the day the point is lost, + with generic solutions efficiency often suffers as well.

    The other issue with generic solutions are that they are generic , ok in JAVA it may well be easier to work within these frameworks - as writing the raw code alternative is a hefty task, not so in PHP, C# with its page-object requirements may not prove very efficient for many many simple situations with the same complexity creeping in at some point.

    I must have written 5 or 6 frameworks for my projects, each supposedly a cure-all for futire projects, but I soon discovered that adapting them for my changeing needs actually took longer and produced many more a dodgy hack than starting from scratch using my own library of goodies.

    PEAR mmmmm changing thoughts on PEAR these days, if you think outside of JAVA and or classical OOP then PEAR can begin to make more sense (PEAR:B excepted) than its given credit for (I think... still arguing this one with myself)

  17. #17
    SitePoint Guru
    Join Date
    Nov 2002
    Posts
    841
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: It seems as though everyone has their own framework

    Originally posted by shoebox
    This also makes me wonder if .NET framework has an advantage (market-wise) since it is a standardized and widely accepted framework.
    It does because it was written by a single entity (sort of) who can keep it coherint. Java has that to a certain extent, too. PEAR lacks that. PHP lacks that. Just look at the sheer number of different string and array functions in PHP. Yet, its taken until PHP 5 to get a decent object model for the language. The strength of PHP is in is price, ubiquity, and specilizations to the task of writing web applications. ( echo "$var" type syntax, $_POST[] vars, setcookie functions, etc).

    Originally posted by mojavelinux
    Right now I am writing a new framework. You might say, great, another one. But no, I am modeling mine off of Struts, as close as I can get without trying to make PHP something it isn't (alot of the java code I am just tossing to the side since it deals with type conversion and all this business that php doesn't have to get involved with). Yes I know about phrame and yes it is pretty good, but I am hoping to go to the next level.
    I am very interested in struts. I have not finished studying it, but I am forming some preliminary conclusions. (anyone know of a publicly downloadable full scale application using struts?) I am not sure about a straight port of it though. I haven't looked at it enough to fully articulate this, but my gut feeling is that struts capitalizes on some strenghts of the java model. Porting struts would mean implementing it in an environment which does not necesarily have those advantages. I may be able to say more on this after some thought.

    If you want to use struts, why not just use java?

    What we really need is a framework that is built for PHP. One that capitalizes on the unique strengths of PHP.


    On another note, the best frameworks that I have seen in my career have been built by taking a bunch of applications and factoring out the common parts, and only adding things when they become necessary by a specific demand.

    The frameworks that I have seen that suck (ala PEAR) have been created by taking a concept forward, building some classes and then extrapolating all possible uses of it without the feedback of use in actual applications.

    For those writing frameworks my advise is to pick a couple of full scale apps that you did not write (maybe PHP-Nuke or phpmyadmin) and try to port parts of them to your framework as you are developing your framework.

    Build your framework by working backwards from existing final code. For frameworks, the simplicity of the application code that uses it is the judge of a good framework.

    Take some tasks from real applications and imagine what framework would have to exist to make those tasks trivially simple. Then, build the framework that does that. Move on to a new task. adapt the framework. after awhile, your framework will be refined and you will not need to change it a bunch for every new task. As the framework evolves, you may find ways of going back to your trivially simple examples and making them more trivially simple.

    This is a way to keep the goal in mind and avoid getting sidetracked on unnecessary things.

    just my 2 cents

  18. #18
    SitePoint Guru
    Join Date
    Oct 2001
    Posts
    656
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I think there is no such thing as a "one-framework-fits-it-all" framework. A framework is like a template (not in the sense of a HTML-template) for a specific type of applications. For example, you could have a framework for web service applications, one for content management applications, one for statistical applications, etc.

    Developing a framework that is general enough for all kinds of PHP applications is very hard, if not impossible. The best you can do is find some sort of common structure in the kind of applications you develop. For me that is mostly CMS applications, in the broadest sense of the word (lot of people call their news script a CMS, so who am I to call my CMS a CMS ).

    Some earlier frameworks I made were not actually frameworks, but more a class library with some 'glue' code. It could not really be reused as a framework in other applications. I think that a framework is not a framework until you can reuse it for another application without copy & paste hacking.

    Finally, developing your frameworks, even if only meant for personal use, is a good thing (tm) . By doing so, you learn to abstract the structure of your application and classes in such a way that you can truely reuse them. Eventually the structure of the framework will stop evolving and you will have achieved good reuse possibilities, in theory of course

  19. #19
    SitePoint Evangelist
    Join Date
    Oct 2001
    Posts
    592
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Finally, developing your frameworks, even if only meant for personal use, is a good thing (tm)
    Naturally, I completely agree

    Before I started out with the implementation of my own framework, I wrote down a list of requirements. These requirements forced me to take some design steps I wouldn't have taken if the requirements weren't there. All frameworks 'out there' are likely very different, because they are based on different requirements. From this follows that there is indeed no framework that solves all problems, unless someone can come up with a set of requirements that captures each and every problem anybody in the world can come up with. And I find that hard to believe...

    Also, whenever I find a framework that has no clear set of requirements it was based on, I'm hardly interested in it, because it's most likely a mess. For a framework to be any good, it HAS to be based on some well-defined requirements.

    Vincent

  20. #20
    "Of" != "Have" bronze trophy Jeff Lange's Avatar
    Join Date
    Jan 2003
    Location
    Calgary, Canada
    Posts
    2,063
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I think forcing a universal framework limits creativity, people going at things with different directions always helps, we would be nowhere today if everyone just decided to do it the same way as everybody else does.
    Who walks the stairs without a care
    It shoots so high in the sky.
    Bounce up and down just like a clown.
    Everyone knows its Slinky.

  21. #21
    Talk to the /dev/null Theiggsta's Avatar
    Join Date
    Mar 2001
    Location
    Tampa, FL
    Posts
    376
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I think you guys should just switch to perl or ruby and get it over with!

    I already have, I learned both perl and ruby and have almost become disgusted with PHP. Now if you think for a second that im using perl in the form of a cheesy compiled cgi, your thinking soooo 1990's. I am talking mod_perl!

    Oh well, you people and your frameworks!

    Honestly I care less for frameworks as all it will end up doing is limiting your skills and increasing overhead. And by limiting skills, I mean you forget how to write the darn thing in the language itself instead of a framework. I also care less for this OO craze as its nothing but tons of hot air and hype. I can agree that OO might come in handy for *large* projects where you have oodles of code that need packaging. However when you get down to quick and simple apps, its simply too much overhead.
    Aaron "Theiggsta" Kalin
    Pixel Martini
    Ruby and Rails Developer

  22. #22
    No. Phil.Roberts's Avatar
    Join Date
    May 2001
    Location
    Nottingham, UK
    Posts
    1,142
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    How nice for you....

    And. Well done for completly ignoring the plus points of OO that have been raised and re-hashed in every OO thread since day one.
    THE INSTRUCTIONS BELOW ARE OLD AND MAY BE INACCURATE.
    THIS INSTALL METHOD IS NOT RECOMMENDED, IT MAY RUN
    OVER YOUR DOG. <-- MediaWiki installation guide

  23. #23
    ********* wombat firepages's Avatar
    Join Date
    Jul 2000
    Location
    Perth Australia
    Posts
    1,717
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    I also care less for this OO craze as its nothing but tons of hot air and hype
    strange you should paddle OOP and say 'ruby rocks' in the same breath !

    stranger still that you say OOP sucks and then register 'disgust' @ PHP which is more at home outside of OOP at the moment, that it will get more oop from here on inwards will not stop its usefulness as a procedural language, that it more suited to web-development than PERL is not an issue for most.

    ... dont get me wrong I agree, to a point, with many of the anti-framework points you raise in your other thread, but your post above dont make sense

  24. #24
    SitePoint Wizard gold trophysilver trophy
    Join Date
    Nov 2000
    Location
    Switzerland
    Posts
    2,479
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Oh well, you people and your frameworks!
    If you've ever written an app that has two or more "views" controlled by if/else logic and perhaps some user authentication, you're also guilty of writing a framework A framework in my mind is simply some code that "triggers" the execution of some other code.

    Honestly I care less for frameworks as all it will end up doing is limiting your skills and increasing overhead. And by limiting skills, I mean you forget how to write the darn thing in the language itself instead of a framework.
    Certainly this is true if you work in an MS environment. Right now they've got developers creating web services for example, with next to no idea of what web services actually are.

    But if you write your own framework, I don't see how you lose out. Certainly eZ publish, for example, takes away the need to think about many things but as has been said, there isn't a framework that fits all problems - I still have to do some "real" coding from time to time.

    I also care less for this OO craze as its nothing but tons of hot air and hype. I can agree that OO might come in handy for *large* projects where you have oodles of code that need packaging. However when you get down to quick and simple apps, its simply too much overhead.
    What about cpan?

  25. #25
    SitePoint Guru
    Join Date
    Oct 2001
    Posts
    656
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    A framework is not just code that triggers other code, it's more than that. A framework is actually a sort of predefined application structure. When you use a framework, the code structure design decisions have already been made by the framework developer, you don't have to redo that design.

    Building applications with a framework should in theory be as easy as creating subclasses of some framework classes. Those classes will then be used by the framework for the application specific behavior.

    What about cpan?
    CPAN is IMHO as object oriented as PEAR (= not a lot). It is not a collection of classes, but a collection of reusable modules, some of which have a class like interface.


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •