Entries Tagged 'Web development' ↓

I’m feeling a little brain-dead at the moment, so I’m throwing this question out.

I want to store items from a couple of feeds (all different varieties of RSS) in a database. What is the best way to make sure I don’t save the same entry more than once into the database?

As part of a gallery I’m writing (and numerous other projects) I found I needed a library of code to embed all different forms of media in web pages. Not finding an existing library to do this I’ve started to write one. As I want to get the (X)HTML code to be correct so I would really appreciate your input. As far as I know there is no resource listing all the classids necessary in the object tags for different media.

I’ve “thrown up a wiki”:http://peter.mapledesign.co.uk/ideas/pmwiki.php/MediaCode/MediaCode to aid discussion/collaboration. The PHP library I write will be released under the BSD license or one of its derivatives (probably PHP License 3.0), free for anyone in the community to use. I haven’t decided about the contents of this wiki page, but I’m thinking about Public Domain or a Creative Commons license. Just in case you’re worried about contributing!

It would be really cool to make a definitive resource in one place for this - I can’t be the only one who is asked to stick videos in web pages?

For one of my clients I’ve been looking for a very simple CMS(Content Management System). It would be nice for him to be able to create new pages for the site without having to ask me to do so.

I’m not thinking of anything heavyweight. Something with a “Create new page” link, allowing you to open up a form and just type away. It would work a lot like Dreamweaver’s templates (i.e. you fill in the boxes). Textile support and WYSIWYG would be present.

There would need to be a way to upload/resize images and insert them. And a decent menu builder - pages can be hidden from the menu, split into separate menus etc.

Knowing the number of CMS’s out there something must do this already. However the majority try to mould you into do working their way, being contstrained by the tool - which shouldn’t be! I played with XOOPS and Mambo, and neither felt it could be bent easily to my will.

In a way it’s a bit of a Wiki idea, being free to create new pages. Weblogs seem just as bad as the average CMS, as they mould you into entries which unless assigned to a special category will appear in the navigation and front page by default. But perhaps they could be adapted to do what I require? I’ve seen it done with MovableType, but no examples with Wordpress, pLog, Nucleus or similar.

Or am I after a breed of tool that no one else is bothered with?

With 4 weeks of -revision- free time stretching ahead of me I’m starting work on my RSS aggregator again. And have reached the stage where I can actually _put_ data into the database.

If I could find a half-decent RSS parser written in PHP and *not* released under the GPL.

So far I don’t have a reason not to use a GPL’d parser (such as “Magpie”:http://magpierss.sourceforge.net, which seems to be the best available) as I’m writing this script for _my_ own use. However, as I don’t know where this script could lead, I don’t want to be limited by the parser’s license.

So can anyone recommend a parser? A BSD-compatible license would be preferred. Currently I’m looking at using Onyx (and on a related note, anyone know what happened to Ed Swindelles? His website (and hence Onyx) has disappeared), but it would require extra development.

I don’t want to write yet another library - I already have enough half-developed ones knocking around (FormProcessor, and a couple of others I’ve forgotten :)).

I’ve started work on my RSS Manager again, and am now cutting code, which led me to put a _lot_ more thought into what data I was going to receive. Originally I was thinking only in terms of processing weblog RSS feeds, now I’m realising that a more general data aggregator would be a lot more useful.

And deciding on the data format to model the database tables on is proving tricky. I am starting to think that following RSS is going to be too limiting; to follow “RDF (there must be a better tutorial)”:http://www.xulplanet.com/tutorials/mozsdk/rdfstart.php too difficult as it’s extensible (someone *please* correct me if i’m wrong) unless I can create virtual tables for each feed, and map all the data to the display model.

Then I found “Atom”:http://www.intertwingly.net/wiki/pie/FrontPage. Is this the answer? Not sure, is my current reply. Here’s a few questions I haven’t found the answer to yet (anyone else find there wiki hard to find _information_ in?).

* Is Atom a generalised format for marking up any content, or specialised just for weblog content? The impression I get is that is has been designed with blogs in mind.
* Is it a subset of RDF or something separate?
* What do the developers see it being used for?

Data format is an important topic, and one which looks like it’ll hold up my development as I don’t want to continue until I’ve got it right. Below is shown the current database structure, based on that used by “Feed on Feeds”:http://minutillo.com/steve/feedonfeeds/. More fields will certainly appear; I’ve got a few lined up for the @Feed@ table already. Feed items will obviusly have to wait until the data structure is worked out.

!http://peter.mapledesign.co.uk/images/rss_manager_db_model.png!

Having had someone asking about removing the “Powered by sendcard” credit from “sendcard”:http://www.sendcard.org/ recently I’ve been thinking about the copyright/Powered by notices that many scripts display, and the requirements many of them make.

Take “PHP-Nuke”:http://www.phpnuke.org as an example (it was a different one I encountered yesterday, but this one nicely illustrates the problem). PHP-Nuke has copyright and license notices at the bottom of every page on the site. If you wish to remove them the author of PHP-Nuke would like to you to “buy a commercial license”:http://phpnuke.org/modules.php?name=Commercial_License for only $300. It appears that the author interprets this passage of the GPL to include copyright notices in the HTML templates:

bq. You may copy and distribute verbatim copies of the Program’s _source code_ [emphasis added] as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program.

My understanding of the passage above is that it is talking about the source code, and how the credits in the code should remain in place, because they are held under the license and copyright of the author. The visible part of the website _is not bound_ by this part of the website - it’s not the code.

While I agree that it’s better if users leave the visible copyright notice, and understand that in the world of free software selling licenses to remove the copyright is one way to recoup the cost of the time spent on the script (I’ve done the same thing with sendcard - “Yes of course you can remove the credit, but a donation would be very much appreciated”) it is a misrepresentation of the GPL to insist that people must pay to do so.

Plus the GPL itself states:

bq. Activities other than copying, distribution and modification are not covered by this License

It could be argued that a web based script is a form of interactive program, and therefore should output the copyright/license notice on each page.

“phpbb”:http://www.phpbb.com agrees that it is within your rights, and have an article in their “knowledge base”:http://www.phpbb.com/kb/article.php?article_id=8 on the subject. It is good to see that they understand the GPL doesn’t cover the _visible_ copyright, but protects the copyright notices _in the code_; but I would appreciate a more lenient stance over moving the notice to a separate credits page. I mean, most desktop software (on Windows) has the credits hidden in the ‘Help->About’ menu, not appearing as soon as you run a feature that uses someone else’s code, and I like to group all credits for a site together on one page - it just feels neater, and IMO makes the site look more ‘official’ and not just tacked together from free scripts like any script kiddie does.

Why has this situation come about? I believe it to be because of the one feature so common in the open source community (no, I don’t mean poor documentation): ego. The desire to be known, to have everyone praise you for what you have written, to have hundreds of people using your script. The best way to do that is to have a visible link to your site.

Here endeth the rant.

Shortly the physics section of this site will start to appear… once I can find a tool to suit. I’ve been considering starting a physics wiki for a whileand the revision I did over the holidays has convinced me that I need a centralised repository for the information I’m learning/finding.

Yet a couple of things annoy me in particular; one exclusive to wikis and the other a much more general, yet relevant problem.

The wiki problem is the bashedTogetherWordsForLinks they use. Uugh. I hate that. Do any of the wikis do it some other way?

But an even bigger problem is data lock-in. Assuming that I write some decent stuff in the wiki, I’m going to want to share it, print it, heck maybe even distribute it. Which means PDF export or similar. Especially if a wiki is to be used as a collaborative writing tool, this is very important. For example, I was planning to collaborate with a couple of people to write a booklet. Now while we oculd all hack it around in Microsoft Word or similar, why not use a web-based tool, and then allow people to browse it while we write (and correct the mistakes)? That hasn’t happend so far - because I couldn’t find the software to do it.

And this seems to be a general problem across most web-based software. People are collecting together so much information and knowledge, yet I see it running into the same problem as the “Digital Domesday Book”:http://news.bbc.co.uk/1/hi/technology/2534391.stm. Although the data will still be readable, it will be in the wrong format, and therefore useless without much work to extract it. And 20 years on it might not be readable.

We need to look forwards while considering data management. Otherwise we might as well not bother to compile and store any information digitally, if it’s going to be lost. It is mainly for this reason that I don’t see paper publications disappearing in the next 30 years.

Yet again I’m asking for suggestions and moaning. Hopefully soon I’ll write some more interesting content in this weblog - like the solutions to the above problems

Found in “Opera’s”:http://www.opera.no changelog for 7.2.0 - “The Regex Coach - interactive regular expressions”:http://www.weitz.de/regex-coach/.

bq. The Regex Coach is a graphical application for Linux and Windows which can be used to experiment with (Perl-compatible) regular expressions interactively.

The tool is written in Common Lisp - this is the first application I’ve run across that (told me it) is written in this language.

bq. It might be worthwhile to note that due to the dynamic nature of Lisp The Regex Coach could be written without changing a single line of code in the CL-PPCRE engine itself although the application has to track information and query the engine while the regular expressions is parsed and the scanners are built. All this could be done ‘after the fact’ by using facilities like defadvice and :around methods.

Sounds interesting

Via “Pseudo Design”:http://www.itnextgen.net/pseudo/, tip number 5 From D. Keith Robinson’s Gorilla Web Tips: “The Myth of Perfect Web Design”:http://www.7nights.com/dkrprod/gwt_five.php

A good post explaining why you should not aim for perfection in web design. Something I need to learn (the name of this weblog wasn’t chosen randomly…)

It’s brought me back to my original goals. Recently my idea of web design has drifted away recently towards aiming for beatiful sites (like “this”:http://mezzoblue.com/, “this”:http://www.stopdesign.com/, “this”:http://exp.hicksdesign.co.uk/ or (_nearly_) any of “these”:http://csszengarden.com/) and forgetting more about the content and usability of my sites. Anyhow, I’ve sat looking at all these sites; then fired up “Fireworks”:http://www.macromedia.com/products/fireworks/ and spent the next couple of hours working on a design. And I _can’t_ do anything like theirs. These guys have talent. (which does annoy me, even though it shouldn’t)

Which leads back to the perfectionism thing. Unless I settle for only doing a good/very good design I’ll never get a site finished. I’ve got to settle for my designs lacking the polish of those linked above, and instead use my strengths to build great sites.

Enough winging and introspection. Time to get back to doing something constructive like programming!

(memo to self: _stop moaning in this blog. Get on and be proactive instead. And remember, you can’t erase an entry from the history of the web once published, so watch what you write. Don’t want it to come back and bite in 20 years time…_)

*Update*
Paul Scrivens “sums up the frustration”:http://www.itnextgen.net/pseudo/archives/our_thoughts/better_days_ahead.php so well in his post, and suggests a solution: _be patient_.

I’m looking for recommendations here. I have already read recommendations for books by C. J. Date and Codd (I think it was); but I’m wondering if there’s any online references/tutorials/articles or books that someone can point me to. I’m a two and a half hour drive from the library with these books, and I don’t want to buy either without seeing them first, hence the request for online resources. And if the online ones are free, so much the better
Thanks all!

h3. Update

As I find resources or are given them, I’ll add them to this post.

* “Visual Case Tool - Database Design Tutorial”:http://www.visualcase.com/tutorials/database-design-tutorial.htm is very basic but gives a good overview of how foreign keys are used.

* “SQL by Design: How to Choose a Primary Key”:http://www.sqlmag.com/Articles/Index.cfm?ArticleID=5113. Another useful article.

* “Devshed: Cross tabulations with MySQL”:http://www.devshed.com/Server_Side/MySQL/MySQLWiz/