September 4, 2008
In order to resolve a potential security issue with the administrative interface Stew and I did a rebuild of the dataset by:
1) shut down the exist server (had to kill the process as there was an I/O error to do with xmlrpc)
2) remove all the files from /exist/webapp/WEB-INF/data/ (I moved them to the homedir)
3) restart exist
4) set passwords
5) restore data from backup
6) use website
September 3, 2008
Checklist Item 1.2: Have done nothing further on fixing the non-valid img tags on the banners.
Checklist Item 5.2: The teacher's notes and supplementary material for the nine webquests are pdf files only. Will get the .doc files from the author, generate html from those and see if that html validates. Then will decide how to proceed.
Checklist Item 6.0: Dublin Core. Confirmed that semicolon is separator for keywords within one subject element. Waiting to hear from Merna on scale.
Suggestion 1: disallow user from seeing directory listing for those directories which do not contain an index file. Done by putting an IgnoreIndex * command in an htaccess file in the root of the site. Greg helped.
Suggestion 2: css link elements need to have type attribute. All pages in the redpath site had this problem, so did a global search and replace on the production site. Kept copy of original site on server temporarily. Not sure best way to do mass updates on the development side, as the page contents are stored in an sql db and generated into files.
Suggestions 3: they don't like duplicate text in "title" and "alt" attributes of img tags. Likely will take no action
Suggestion 4: broken links. After considerable testing, it appears that the cybermysteres.ca domain points to the wrong place (and not the same place as the mysteryquests.ca domain). Unable to log in to the control panel administrative interface to confirm for sure. Greg and Martin helped indendently. There were also a handful of typo errors in links which I fixed manually on the production and development servers.
Met with JS-R to discuss future plans for the CityStats.ca project. This is the outcome:
- A duplicate development site will be created, where extra data can be added for research purposes, and where development for the main site can be piloted. (This is now done.)
- A more sophisticated user interface will be created so that visitors to the main site can create sophisticated datasets through a wizard, which will generate tables. I've already completed an outline of the wizard interface and functionality; this will need to be elaborated with plans for the AJAX front end (probably based on JQuery) and the PHP backend (planning for which will wait until some coding changes currently being done by JD are complete, and integrated into the current site).
- The current site will be reviewed as part of the review of a paper being considered for publication; feedback from that process will be discussed and factored into the plan for the main site.
- I'll set aside some time in October/November to work on the changes to the main site.
- A fully-functioning wizard/table/chart system needs to be in place by January, when the article is published.
Started work on response to "Technical Requirements Evalutation Checklist" for the last three Canadian Mysteries sites (Klondike, Jerome, Vinland).
Checklist Item 1.2: The banner image for the 9 webquests do not validate, but problems are trivial. I'll need to spend a couple of hours to make those changes.
Checklist Item 5.2: The teacher's notes and supplementary material for the nine webquests are pdf files only. They need to be provided as HTML as well. Each webquest has one text document and 1 to 2 documents with tables of text (evaluation guides for teachers) and 2 to 3 documents with empty tables (for students to fill). This will take a couple dozen hours to create the templates and then maybe 80 hours of copy and paste of text from a pdf editor.
Checklist Item 6.0: Dublin Core. The keywords in the the subject element should be separated by ";" not by "--". I've asked Merna how many pages contain DC elements.
Suggestion 1: disallow user from seeing directory listing for those directories which do not contain an index file (need to do research).
Suggestion 2: css link elements need to have type attribute (need to figure out scope of problem then do mass search and replace)
Suggestions 3: they don't like duplicate text in "title" and "alt" attributes of img tags. (need to figure out scope of problem and ask Merna for direction)
Suggestion 4: broken links. I'll fix those.
The Philosopher's Annual aims to select the ten best articles published in philosophy each year, "an attempt as simple to state as it is admittedly impossible to fulfill".
Against a background of twenty-four volumes in hard copy, the Annual is now available entirely online at http://www.philosophersannual.org/
September 2, 2008
Abstracted a new item function to make it usable for both keywords and projects, and then hooked it into the GUI. Got it working in one of four places (the keywords tab), and half-working in the Projects tab (adds data OK, but returns empty option elements). I should have time to iron out some of these issues tomorrow; it's probably something to do with JavaScript. We'll get there...
Many fixes, and some questions still open for editorial decisions.
Just when you thought you had enough browsers to choose from (IE, Firefox, Safari, Opera, Konqueror), Google enters the fray with Google Chrome (currently only for Windows, and, naturally, it's a beta).
This rather impressive comic (http://www.google.com/googlebooks/chrome/) introducing the browser and the rationale for building it is really worth reading, and seems to show that they're doing things slightly differently from the other browser builders, with what seem to be solid decisions on the speed and security front. But having played with it, there are a few things to dislike about it.
For one thing, I can't find any way to put a normal search box on the GUI, so you're forced to use the blended search+URL bar, which I find a bit confusing. More seriously, the cookie handling is rather scary. These are the options:
- Allow all cookies (the default)
- Restrict how third-party cookies can be used
- Block all cookies
In other words, there's no option to have the browser ask you whether to save cookies for specific sites.
The about:memory URL shows the kind of thing they've been focusing on (small memory footprint); it shows you the memory profile of all the browsers you currently have running, including IE and Firefox.
It also has the same bugs as Webkit/Safari (predictably), especially with JQuery UI components. There's no option to change the UA string (as far as I can find), so debugging these things on sites which already browser-sniff to work around them is going to be difficult. But it does have a neat little reporting mechanism for buggy web behaviour; it'll take a screenshot of the current page and send it off to Google with your explanation and the URL, automatically.
My biggest hope for this is that Google will put some developer time into bugfixing in the Webkit engine, which will help clean things up for Safari (and possibly Konqueror) as well as Chrome.
From the New York Times:
At an experimental Web site, Many Eyes, (www.many-eyes.com), users can upload the data they want to visualize, then try sophisticated tools to generate interactive displays. These might range from maps of relationships in the New Testament to a display of the comparative frequency of words used in speeches by Senators Hillary Rodham Clinton and Barack Obama.
The site was created by scientists at the Watson Research Center of I.B.M. in Cambridge, Mass., to help people publish and discuss graphics in a group. Those who register at the site can comment on one another’s work, perhaps visualizing the same information with different tools and discovering unexpected patterns in the data.
Visit many-eyes.com
Read full article