A little update of the elk_mediaplayer plugin, which was also the occasion to update to the latest version of JW mediaplayer.

Please note that as the pages I was using it on have moved from Textpattern to MODx and from JW Media Player to Flowplayer, I won’t be able to update it any longer. Anyone interested in maintaining it is very welcome to do so.

0.3 update

Thanks to Steve, the version works with JW FLV Media Player 5.

New variables:

• playlist (over, bottom, right)
• playlistsize (number; replacement for plwidth)

Made XHTML 1.1 compliant by replacing commas in id="v_1,2,3" with underscores: id="v_1_2_3".

To-do: Playlist images implementation.

elk_mediaplayer_v0.3

It’s been a while since I have posted a little animation from the steps of my sketches (that I sometimes save regularly enough to give an idea of how it works), so here is an update. Other previous time-lapse videos can be found at my website tagged “process”.

So I’ve updated my youtube page with this little video… Perhaps I’ll try vimeo also some day. Enjoy!

EDIT: added it on vimeo too

ScreenGrab! A very handy Firefox add-on to make quick screenshots of the pages you are browsing. It can grab a selection or even a whole page scrolling down to a few pages and save it to the clipboard or a nifty PNG.

And while I’m having fun with Wordle, I might as well try it… it works well with Java applets, flash and the likes

As I was publishing a new version of my website, I noticed that some of the little animations I had previously inserted with the bas_flvplayer plugin had ceased to function. Well, apparently… because on other computers they seemed to still look fine.

I noticed that it had stopped functioning for my browsers with the latest Flash version (10) and after some research, it appeared that Adobe has decided to implement a few security “improvements” on the way things are allowed or not to be loaded by Flash.

The thing is that the plugin calls the files in a “download” mode which changes the headers and streams the files. While I didn’t manage to tweak the crossdomain.xml file appropriately to have it work again, I changed a few lines in the plugin and voilà.

If you happen to have the same problem, you can go for a quick and easy fix to the Admin panel, plugins, and then click “Edit” on the bas_flvplayer line.

I have already added a few mods on the plugin (to add sounds and a more standard compliant insertion of the player) so I can’t tell you the exact line, but if you make a search for the function function bas_flvplayer_get_file($where)

You will find this line

// hu.gTxt('file_download').'/'.$thisfile['id'];

That you can comment out and replace by the following

hu.'files/'.$thisfile['filename'];

Where “files” is your folder’s name (would need to find the textpattern proper call for that). Note that as the flash files now will go through a GET mode, it’s not tracked by textpattern and the download counter won’t increment. (I couldn’t care less, as these files are embedded anyway.)

You have also probably to change the if(!empty($thisfile['id'])) by if(!empty($thisfile['filename']))

And that should be all… Enjoy!

Thanks to the pervasiveness and performances of the Internet, it’s now possible to have very complex programs run from your web browser, programs that were only destined to your desktop before.

It’s usually called “the Cloud” (or cloud computing), and allows the access via the Internet to complex programs. You probably already knew about Google docs and your favourite word processor, or spreadsheet program, but coming from up the cloud, we are talking about different kinds of birds. Graphic programs more like Adobe’s Photoshop, except that here, it’s for free (at least for the basic version), and you don’t have to install anything on your computer.

Check out Aviary; even if you’re not interested in computer graphics, the website concept alone is worth checking out, with each project being represented by a colourful bird, the symbols’ continuity being pushed to having eggs for their coming projects (which look very interesting as well), or a broken egg for “404 not found” errors.

The latest of their release is “Raven”, a vector editor which even if quite simple compared to Adobe Illustrator or its open-source counterpart Inkscape is already quite promising.

One can imagine many more interesting ways of using this kind of program, especially on the line of collaborative work. Up to now, there was the early versions of OpenCanvas which allowed for instance sharing a common canvas and draw pictures at the same time with different individuals, but now there seem to be a wider range of possibles.

Here we go again… Another hermetic RIP failing message from Lulu.

180: Error- Text block will not RIP - CID 5722427

I checked again my file, and corrected what may be the cause of this. As Lulu’s policy now seems to be to directly refund your order and not bother with detailing what may be the causes of the PDF failing to RIP at the printer’s, I thought it may be good for me to write down the major reasons I’ve found that may be prevent a proper print job at their printers’.

Before I list them, some tips how you can “debug” your PDF (from the latest in the process, to the earliest):

• You can use Pitstop Pro free 30 days demo; it has a selection feature which highlights nicely the bounding boxes and makes it easier to look for content extending from the crop area.
• You may use the preflight options (of Acrobat, or your DTP software); there are useful diagnostics you can find there, deep down in the options
• You can do an EPS export first of your pages. It’s long and tedious, but if you open the EPS in an appropriate software, you may notice quirks in the export before (how the page was flattened for instance; sometimes an intermediary high-resolution export of the little complex transparent illustrations to a PNG/TIF format may improve greatly the complexity of a page, and make the RIP job less bogged down at the printers’)

Now, some of the main pitfalls:

• CID encoded fonts — avoid them; perhaps not the reason for that message (since I got it after vectorizing all the faulty characters, but not taking any chances, they were displayed strangely in the EPS exports anyway).
• Images extending outside of the page canvas. Don’t believe the PDF crop, some may still be hidden around. The main reason is still the facing pages (I work in facing pages to have a proper page numbering and composition, and it’s not a problem as long as you export in single pages, and check out for those overlapping contents).
• Fancy effects on the texts: they usually create some “text used as clipping path” which doesn’t go too well in my experience. Better rasterize if absolutely needed, or find another method.

Here follows a short list of plugins I use with my installations of textpattern (TXP) that I find difficult to work without. (They are essentially to make the life of the webmaster easier, the features are not so much user-oriented).

ied_plugin_composer, by Yura Linnyk

Allows you to edit your plugins, export them in the encoded form used by the administration interface, etc.
Very useful for most plugins that don’t require external files to work.

rss_admin_db_manager, by Rob Sable

Allows to manage your databases inside TXP; a light-weight phpMyadmin. Extremely useful when you are coding a plugin and need some specific field, or when you want to check something inside the DB (size, etc.)

Even more useful is the integrated backup management; you don’t even have to dump your database manually, it’s done with one click, and very intelligently stored in the file system so that you can restore a previous version with just another click.

mcw_templates, by Mike West

A plugin which allows you to export/import your system’s look and feel (forms, css etc.) in files. Pretty useful when you juggle in the beginning with various themes to test your system.

tru_tags, by Nathan Arthur

A very complete and well-thought tagging plugin. Allows you to manage your tags, insert them easily into your posts.

bas_lightbox, by Bastian Sackermann

Allows the integration of the lightbox unobtrusive javascript image display tool.

asy_sitemap, by Sencer Yurdaguel or jmd_sitemap, (updated version) by Jon-Michael Deldin

Generates a gzipped xml googlemap of your site

Wisp — volume 2 / front cover

After having published with much satisfaction a few books¹ with Lulu (one of the main actors in the print-on-demand services) this post is listing some of the things that may be useful to know (or to remember after a bit of trial and errors) to design a source file that will produce the perfect PDF file.

Black and white prints

Not much to say here, the quality was correct, and the first PDF I created went without any problem.

Colour prints

A bit more of a headache at first. The PDF produced for the book(s) had intricate layouts, and when produced according to the specifications (full embedding of the fonts, flattening of all transparencies etc.) it failed to RIP properly.

Thanks to the Pitstop Pro trial version, I was able to see what the Lulu contact indicated me, that some of the contents were going off the printable boundaries of the pages.

Then I proceeded to slice all of the stuff that I had allowed to go out of the pages without paying too much attention (in the past, I’ve worked with printers who never had any skirmishes printing files created that way, but I figure this was not the case with Lulu’s printers).

Later, I discovered that this happened when creating the PDF, due to the bleed settings in facing page mode, when you export each page individually, some of the parts bleeding on the next page were not necessarily properly cropped; most of the times they are just copied, and “hidden”.

In fact, I figured out that even though working with the bleeding settings was very convenient (to flip between the final page preview and the working page), it’s a real pain when it comes to exporting the pages produced that way. Why? because when you work in facing pages, the interior bleed (of say the even page) takes into account the page next to it (the odd one) –while you don’t really care, because these parts usually fall in the gutter. So you never can guess how the export will crop your page (esp. if you get vector art or text intersecting those).

So, a few things to make sure it simplifies the creation of your PDF for Lulu:

• don’t use the bleed settings: put the bleed to zero, and for full-bleed, extend the page’s size to the desired dimensions: for instance, for a US letter (8.5“x11”) format, your working file will be 8.75“x11.25″. You can adjust the bleed markers by using the grid and guides.
• When you compose the file, make sure your backgrounds, images, and esp. vectors and texts do not extend outside of the page. If need be, do not rely on the program to crop it for you at the export if you think reducing the box’s boundaries will be enough. Most of the times, even with the cropping option activated, it doesn’t work that nicely on the exported PDF.

But this was not all; after resubmitting a revision of my PDF, it still wouldn’t work. Various reasons were invoked (fonts embedding, too many colors spaces, or ICC profiles) but as far as the preflight was telling me, everything seemed nice and Adobe 5.0 (PDF 1.4) compatible…

So, another reason was invoked; that some areas where more than 600 dpi and cluttered the print job… Okay, I go check and yes, some areas where the transparencies flattener had been over-zealous were a little above 600 dpi. I adjusted the PDF to no avail.

So, finally, I decided to redo the file by exporting most of the complex backgrounds to EPS (without the texts), checking the vectors, and exporting them to TIFF at 300 dpi, and reimporting them in my file.

I took care of getting rid of the transparencies (in InDesign) which are convenient, but wreck havoc on the PDF. Going to plain simple PDF, like in the dark times of good old un-fancy Quark Xpress.

This process finally made the PDF printable, and I figured out that the mysterious “Text won’t RIP” message I kept getting from the printer was probably due to those multiply/drop shadow transparencies/effects in InDesign. Because checking with the preflight the PDF the option “text used as clipping path” gave a few warnings in the initial files for all the texts using the fancy effects. They had disappeared in the simpler version.

To this date, there still seems to be a little issue with the printer’s way of cutting the pages… To be checked.
[Note from 2008/11/17: after a re-order was made, the quality was far better, and the page trimming was what it was supposed to be; I suppose the first shot was just messy as these things can happen]

So, in short, if you plan to make a full-bleed full color print with Lulu, remember that simple designs are the most effective, and if you want to go for something more elaborate, make sure you don’t rely on the PDF exporter to create a nice PDF; you’ll get better (and quicker) results by simplifying it for the program.

The silver lining is that the efforts you’ve made in simplifying the layout will make the PDF export much quicker!

Endnotes:

1 First of these babies, Circle of Eights, a thick (350 pages) book, printed in black and white, without full-bleed. A compilation of collaborative stories which have been running for more than a year.

Other books are Wisp (volume 1 and volume 2) a quarterly compilation of the Wisp ezine printed in full colors, with full bleed.

Greasemonkey is an extension for Mozilla Firefox that runs user scripts on pages to modify their content, rearrange them, add functionalities etc.

It comes pretty useful in websites like Flickr. For instance, two scripts that I found really interesting:

• Flickr No Awards — To hide the nasty defacing awards in the comments of popular photos. The nice treat is that it’s possible to reveal them if you want to.
• Flickr Hide Comment Images — Zaps the images off the comments and photo pages. Actually, I use that only on the recent activity pages, and not on the photo pages, were the first one does what I want.

After a long wait, Firefox 3 has been released officially, following release candidates that were available for a few days already.

The Mozilla team has been trying to setup a first world record of number of downloads in a single day (June 17^th), and seem to have exceeded expectations with more than 5 millions downloads so far…

Flock is on its way to incorporate the new version of the Firefox engine at its core, and has released its version 2 beta. (this post is written using this beta version)
There seem to be some real nice memory handling and performance improvements, and that’s really all that I’ve been anticipating from this new version.

So far so good