Related
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
I found around 8 Strenghts and flaws of using APEX over another program(http://en.wikipedia.org/wiki/Oracle_Application_Express), but i am not sure i quite understand WHEN to use it.
From what i understand, if you want a fast and easy-to-use development tool related to Oracle, Apex is your first choice. While if you need a complex solution, APEX won't fit.
I would like to know you guys opinion on this. In my case, i need to know if i should recommand this product or not for a raquetball club. Since it is not a big company, i believe HTML DB would be the best choice because we want as less manipulations as possible once it's implemented. We also don't want the owner of the club to pay a lot of money to get somoeone who can develop updates.
Apex is free, and an oracle XE database is free too. Apex is rapid development.
But what you're asking depends on so much more.
From what i understand, if you want a fast and easy-to-use development
tool related to Oracle, Apex is your first choice.
Is an oracle database already being used?
Easy to use dev tool: yes, sure. But as with anything, it depends on whether you have some experience with it, what the specs and expectations are, ...
While if you need a complex solution, APEX won't fit.
Well,... Would a complex solution be so much less complex in another environment? Just how complex are we thinking? In my opinion, you can go pretty far in apex and adapt it to your needs. it might involve creating templates and plugins to set up a framework, but it is doable. An example would be apex projects whom have been completely integrated with ExtJS. Apex is not the answer to everything too, but it's good. If you'd stay within the Oracle stack for more involved/complex, i'd say the next thing is ADF. Personally, i'm not convinced by that one though. It also has its pros and cons (such as: requiring java knowledge. pro for some, con for others...)
In my case, i need to know if i should recommand this product or not for a raquetball club. Since it is not a big company, i believe HTML DB would be the best choice because we want as less manipulations as possible once it's implemented.
How large is this club? How intense will the website be used?
Is there a DB already? Is it Oracle?
Who has DBA knowledge (even basic)?
Who will develop?
If your specs are up to spec, then much fiddling shouldn't be necessary after the launch.
We also don't want the owner of the club to pay a lot of money to get somoeone who can develop updates
Who will host the server? Who will run it? Who will administrate it? Do you plan to go cloud-based?
I'd almost ventured to suggest PHP may be a good alternative if this is a small project. Those developers may be easier to find and less expensive than an apex/oracle developer. But then again, if you're planning to outsource it may be less of an issue. If your oracle instance would be in the cloud somewhere, you'd even be pretty safe i'd believe...
Really, what options are you trying to compare? You're asking about apex, do you have any experience with it?
Honestly, your question is not so much a question as it is an opener to a discussion. Each technology and database will have its pro's and cons, fans and dislikers.
Personally, i really like apex. It has a lot going for it, especially when you're already invested in the Oracle stack. And it is still growing, getting good support, and great releases with lots of new features.
I can't really say how it must be set up a (reliable) service from the ground up: server and database, doesn't matter how small, you'll need some understanding and knowledge for that. If you just wing it and cross your fingers, you'll burn them somewhere down the line. Same with development. But as far as i'm concerned that goes for any other tech. Unless you outsource those aspects of course.
Etc etc. There is much and more to be discussed this way.
I put a downvote there for these reasons: there is much to be discussed, it isn't really much of a question which has a conclusive answer. And it maybe only could fetch a conclusive answer if you put more specifications up there and get people of the concurrent platforms to respond.
Edit
I'd like to react on Daniel's post.
First I have to say that I am originally a PHP-Developer, and I really like this language and environment. Nevertheless I decided to do an internship this summer, where I am currently working with APEX. Together with another intern I am developing a bigger application, and I hope to give you some useful input. This only covers the development of the application, as I am not really involved in things like database administration and so on (although I have to say that a PHP Webspace with a MySQL Database also isn't to hard to administrate, especially if there are not too many users).
I'm a developer too, and i don't do server and database administration (corporate environment). Not that i haven't dabbled a bit here and there in my spare time, but i digress.
But my experience is that it gets very, very hard to solve tasks which are out of this range. That doesn't mean that you can't get multiple tables in one reports, some joining of tables is also not a big problem, and can be easily achieved. But if your application needs even more than this I cannot recommend APEX, so I totally agree with your rule, that complex applications should be created in another way.
So speaking as a PHP developer, would you even recommend PHP to achieve this? Would it be less or just as complex?
I'd also argue about complexity. I've worked on some large forms, which for me by now means there are more than your average amount of items, some dynamic actions, validations, maybe some custom process(es). I've not encountered extremely complex situations and honestly i'd question who created those expectations. Thinking outside the box may be a virtue at times, but that doesn't mean the box is always bad. Complex mechanisms or pages can maybe be broken down into more easily accomplishable parts. An example would be using modal pages to break it up.
I also think that the slogan with the limited programming experience is only partial true. It is only true if you have only easy applications, as you have already said. I personally also can't stand the mixing between an IDE and "easy to use"-forms.
I agree about limited programming experience: you'll only create the most basic forms and reports, and having almost no experience i'd think you'd shy away from even option pages in fear of breaking something. Same thing goes for even basic db and server administration: i wouldn't like to rely on such a person when there is no experienced backup (but a very small project as is apperently being described might be acceptable).
I too am only invested in the programming side.
It's also not very easy to create new templates and other things, at least in my opinion it would be much easier with other frameworks.
I'd say that the templates make things very flexible and are certainly not hard to use
And maybe even the worst thing: I think that this application is quite buggy. I don't know how many times simple deleting and new creation of a process/page/validation or whatever solved a problem, where you are simply not thinking of this solution.
Wow. I strongly disagree with this. I've maybe encountered one such thing over the course of a year and tons of forms. Not that i haven't heard of some problems on the OTN forums, but usually they had to do with upgrades.
Summary: I would only use APEX if you have application which REALLY fits it. That means just reports and forms, everything which is more results in pain debugging sessions (as this is also not easy in this environment...) and bad maintenance.
It is too bad that there are not that many large, public sites which use apex out there. As far as i'm aware, there are and have been large project involving apex, but those usually are in a corporate environment and thus are never shown off (can't be). I personally believe apex can be pushed a lot further than the basic forms+reports you mention (and i mean basic, because things will usually come down to forms+reports in this context).
Debugging does not have to be a pain though. If you provide enough debug messages and comments in your own code, that will help a great deal. Debugging the page, javascript console, and if needs be an autonomous error logging procedure to be used in your plsql packages,... I'd say there is plenty to help out (and if you're driven to this, you are working on some more complex material, and i assume that you have the knowledge to actually deal with the complexity you've set up).
And interesting point you raise lastly is maintenance though. I'd say a point on which apex should improve a lot is versioning, out of the box. Exports need to be improved so they can be broken up a lot easier.
Wow, look at that wall of text... I could've guessed this would turn out into a discussion.
First I have to say that I am originally a PHP-Developer, and I really like this language and environment. Nevertheless I decided to do an internship this summer, where I am currently working with APEX. Together with another intern I am developing a bigger application, and I hope to give you some useful input. This only covers the development of the application, as I am not really involved in things like database administration and so on (although I have to say that a PHP Webspace with a MySQL Database also isn't to hard to administrate, especially if there are not too many users).
To start we created a few applications, just to get a feeling. Afterwards we started with the easy parts of the application. APEX is really great if you only have to build some reports and forms to edit the entries of the database. It's very fast to create these things with the integrated wizards.
But my experience is that it gets very, very hard to solve tasks which are out of this range. That doesn't mean that you can't get multiple tables in one reports, some joining of tables is also not a big problem, and can be easily achieved. But if your application needs even more than this I cannot recommend APEX, so I totally agree with your rule, that complex applications should be created in another way.
I also think that the slogan with the limited programming experience is only partial true. It is only true if you have only easy applications, as you have already said. I personally also can't stand the mixing between an IDE and "easy to use"-forms. It's also not very easy to create new templates and other things, at least in my opinion it would be much easier with other frameworks.
And maybe even the worst thing: I think that this application is quite buggy. I don't know how many times simple deleting and new creation of a process/page/validation or whatever solved a problem, where you are simply not thinking of this solution.
Summary: I would only use APEX if you have application which REALLY fits it. That means just reports and forms, everything which is more results in pain debugging sessions (as this is also not easy in this environment...) and bad maintenance.
I am posting here as a guest--hopefully, I will create an account. I used O-HTML-DB from its early releases. We built pretty fine apps. I last used it in 2004, having moved to non-development roles.
Since 2007 though, I decided to revisit the tool and found out about APEX. I have since had my own test apps. I disagree with most of what the intern says.
If you have a limited objective (your business need and associated requirements), then your use of APEX will be limited. This is a very robust application, with sophisticated security features (I have been in InfoSec/Cyber since 2009.
Yes, you are correct than claims about APEX not requiring a solid development background are not accurate. You need to have a sound grasp of SQL/PL/SQL, JavaScript. But you c an also take great advantage of OTN, where developers generously share their know-how. When I started with O-HTML-DB, I had never built a DA in a business environment before. I had theoretical SQL skills. I was a Web Developer with a grasp of JavaScrip, training in Java from Learning Tree International (in addition to academics). But I and my colleague (we were two developers) learned a great deal from OTN. We built three Web apps, one of which supported over 6,000 users--just O-HTML-DB!
APEX has taken O-HTML-DB to new dimensions. You do not need to hard-code validations like we did with O-HTML-DB. Of course, you can modify, which requires a sound grasp of SQL/PL/SQL.
Maybe templating is somewhat confusing to many developers, understandably. But as you continue to "play" with APEX, I am sure you'll like it. It can do nearly anything. Only your limited vision will restrict it.
Erick
Can anyone expound on disadvantages, if there are any, to using a ColdFusion development framework? I'm developing an application traditionally, and I'm tempted to use a framework having seen how simple some things can be done.
I'm new to ColdFusion and frameworks in general. I want to understand the implications of using a framework, including advantages and disadvantages.
Disadvantages:
learning curve (pick a lean framework to reduce this)
front controller makes ugly URL, often needs URL rewrite on web server layer
risk of framework being discontinued (no support, hard to maintain, break on new CF ver)
framework bugs (pick a popular framework with good & fast support)
harder to debug sometimes, since actions are generally not a .cfm anymore. Tip: make use of cfdump and cfabort to see the dump in the controller layer
some frameworks takes longer to reinit. Since most frameworks will cache the configurations and controller layer for performance, during the development phase, you'll need to reinit all the time. CF9 eases this problem 'cause it is much faster.
lastly, sometimes you'll be using framework's API, an abstraction from CFML, and missed out on the native ColdFusion way of solving the same problem.
Performance generally is a non issue. Don't worry.
Henry's already given a good answer, but I would just like to pick up on this part of your question:
But does it not come with a performance tax?
The performance overhead of a framework is negligible.
In fact, you may even get better performance from frameworks such as ColdBox, which have built-in caching.
Remember, most frameworks are mature codebases used by lots of people - most likely, your newly written untested code is going to be the culprit, not the framework.
However, as a general rule (not specific to frameworks) performance is not a problem unless you've got measurable results that say it is.
i.e. don't just think "I'm going to do X instead of Y because I think it'll be faster" - go with the simplest option that meets user's needs, and only change it if you can prove that it has a performance problem and that your proposed solution is better.
It depends the nature of project you are into. I think its always advisable to use a frameowrk for better code organization, scalability, conventions and other. If you are supposed to start with a enterprise level application then coldbox is the best framework as far as my expriece goes. It has a bigger learning curve but its worth learning. If its simple start up project then FW1 is good. You can find a list here
http://www.riaxe.com/blog/top-coldfusion-frameworks/
Has anyone of you done evaluation of a django project and how to improve/refactor it's code base? A pet project in company I work at is becoming more widely used and it would be good to improve its quality before further development. Are there any techniques or methodologies of analyzing django projects before we start putting more and more features into it? We wouldn't like to suddenly realise, that due to earlier poor choice we have to live with something really bad.
What I noticed after working more than a year on a quite large site was mostly this concerning our design/coding. It's not purely refactoring related and probably you already know a lot, but maybe it can help :).
Most importantly, we did not always put our code in the right place. There was too many functionality in the view methods, too few in forms and models. I see this problem many times. To handle input, use forms, to format/tweak/... model data, use model methods or properties. Seriously, at one moment I reduced a view of about 150 lines to 20 by just putting code in better places. Improves maintainability and readability very well
Many code we wrote actually did not really use the full force of python and/or django. For the first one, reading something like Dive Into Python was a great help, for the second one I just tracked down all complex constructs we built (mostly legacy code from 0.96) and looked if there were alternatives in the django docs. Of course, don't waste your time trying to reduce everything to a one-liner, but certainly with legacy code this can help improve readability and maintainability.
Always look on sites like django-snippets, google code, ... if there are decent existing django projects that can take away a lot of your functionality. Often these projects are looked upon by more people and are therefore more stable and performant. If a project doesn't meet all your demands it can even be a good idea of just adding your requirements yourself instead of doing it all custom for your site.
Try to keep the application cross-over dependencies minimal. When you would draw a dependency graph (e.g. by linking each app that have foreign keys to each other) it should still be clear and not something where every app is linked to any other app. Typically you'll have some 'helper' apps (e.g. user system, tagging) that are used by many and all other apps are actually only dependent on those apps.
Write tests, django has an excellent test suite, so use it. Certainly for parts of code that are common to many apps and are likely to change. Really, nothing so annoying as suddenly noticing a bug you actually solved 4 months ago and you have no clue which update since then broke it again.
Take a second look at database normalization, the django orm model is still tightly coupled to relational databases and so it's an important concept, certainly when you work with models that are likely to be expanded later on.
Concerning real refactoring, the only important tool I can think of is South, can help you a lot if your database scheme changes. Otherwise, like a hint I already gave: write tests to make sure your functionality before and after refactoring remains the same.
You can learn about code smells. Some folks say that if your code smells it means you need to consider refactoring.
This is very wide subject in fact. Remeber that:
Premature optimization is the root of all
evil -- DonaldKnuth
I personally think that this is also valid for (premature) refactoring.
EDIT: this is also a good resource about code smells: http://www.codinghorror.com/blog/2006/05/code-smells.html
This is related to this question:
how-can-i-port-a-legacy-java-j2ee-website-to-a-modern-scripting-language
but with a narrower focus.
We're pretty much rewriting our legacy Java app from the ground up for a variety of reasons, but attempting to keep the user interface pretty much the same.
In one of the answers, someone said:
Pick representative pages and rework them into the appropriate templates. You might make use of some legacy JSP pages. However, don't waste too much time with this. Use the HTML to create Django templates.
The thing is, the application has about 15 different "front pages", each of which is largely composed of an appropriately mixed&matched subset of ~100 jsp files. I believe the partitioning of content in the jsp files is probably the partitioning that I would want to use for the Django templates; thus I'd rather convert them to templates than start with the HTML and re-do the work of figuring out the proper partitioning.
So, I'm really hoping there is a reasonable way to do this conversion.
I'm a Django newbie & have never worked with javascript (tho other folks on my team have), so any help would be greatly appreciated: if you can recommend an automated or semi-automated tool, or suggest a basic approach, tips & tricks, advice, horror stories..
Thanks in advance!
Edited in response to Vinay's answer:
I think it is fairly vanilla jsp, with no third party libraries involved.
There are scriptlets, but they are well-segregated into their own files.
At this point I think we want something quick that we can improve
to best-practices later.
Regarding EXTENDS versus INCLUDE, this makes me think that we ought to
think carefully about the design of our front pages, at least; but can
take a least-effort approach to converting some of these small snippets.
You might be on a hiding to nothing - it'll certainly depend a lot on the details, e.g. your use of third-party libraries and custom JSP tags. Do you just want to implement it in Django with minimal work, or do you want the finished site to reflect Django best practices? For example, in Django it's common practice to design templates using their extends feature - although Django does have an include tag which is analogous to <jsp:include/>, one sees extends more often and include much less often. In the JSP world, the extends functionality is not part of vanilla JSP and is implemented using third-party libraries e.g. SiteMesh or Tiles.
If I were faced with the task of "get something that works - fast!", my first approach would be to work up a Python script which parsed all the JSP files (using a fairly unsophisticated parser based on regular expressions) and spat out the equivalent Django template files, flagging up the hard stuff as it goes. It probably won't be pretty and it won't do everything well, but it'll be systematic and you'll get a checklist to things which need to be done by hand (e.g. template tags which need to be implemented). The likelihood of this approach adding value is proportional to how tidy and systematic the JSP code is, whether you use scriptlets at all, and so on.
What are the smashing (pun intended) features of grok that makes it better than django?
how do I know when my project needs grok+zope, or it can just be developed with django ?
Zope was the first object publishing framework evah, and the Zope community has a long experience with Doing Things The Right Way. Zope 2 was the first attempt, Zope 3 was the next attempt, and we are now into the third generation of web frameworks, which includes Grok, BFG and Bobo.
Grok is massive, and has even more modules available that doesn't come when you install the base (and it's in the process of reducing the number of required modules as well, so the footprint gets smaller). BFG and Bobo go the other way around, and are minimalistic frameworks but with easy access to the Zope Toolkit and all the functionalities of Zope.
And although Django is making many of the same mistakes Zope2 did, they are also fixing them much faster, so I completely expect much of this discussion to be moot in five years, because I expect every single Python web framework to use WSGI+WebOb+Repoze+Deliverance+Buildout as a base by then. But even then I'd go for frameworks where I can use the Zope Component Architecture and ZODB, but that includes not only the ones made by the Zope community, but also for example Turbogears. And maybe it will include Django too by then, who knows... :-)
Depending on what the project requirements are I would today go with either Plone (if they need CMS), Grok or BFG (depending on the involved developers, and the complexity of the task and the budget). This is of course partly depending on my large experience with the Zope technologies and my small experience with Django, but mostly because I can use ZTK and ZODB in Grok and BFG.
YMMV, etc, blahblah.
Zope Toolkit
Brandons talk on the Zope Component Architecture (Video from PyCOn, Slides from PloneConf)
BFG
Bobo
Grok is basically all the power of zope in a way easier to use package. So you do get all the luxury of a real python object database (though you can use an sql backend). And I assume you know about the adapters/utilities/views of the so-called "zope component architecture". Those allow you to make a robust application. Especially handy if you later need to selectively customize it. And security is traditionally a zope (and thus grok) strong point. Development and deployment are handled fully with eggs (and buildout): in my experience this is a robust and reliable and repeatable and comfortable way.
If you have an application that can work with straight sql tables without needing much selective customizing afterwards: nothing wrong with django. You'll have to do much security yourself, so that needs a keen eye. There's much less of a framework behind it (an ORM and a url mapper), so your python will feel more "pure and simple". This also means you need to do more yourself.
There's nothing from stopping you from selectively using parts of grok: http://pypi.python.org/pypi/grokcore.component for instance is very much the core. Pretty well isolated, so you can use it without buying into the whole zope stack. I'm pretty sure you can use that in django. grokcore/zope component is just python code. This gets you the adapters/interfaces/utilities. I don't know what you're building, so you'll have to experiment.
One thing hugely in favour of grok that I'd suggest trying out: zope's ZODB object database. A good ORM (and django's is pretty ok) helps a lot taking the pain out of SQL databases, but a real object database is just plain luxury :-)
I don't think any of the frameworks are intended to have any 'features' that make one 'better' than the other, or 'needed' in certain circumstances. Rather, the difference between Django and Grok - or Pylons, or Turbogears - is really one of approach. You may find the approach of Grok to your liking, or you may prefer one of the others. I doubt there is much you can achieve in one of them that you can't in any of the others.