software

You are currently browsing articles tagged software.

William borrowed my camera to go on his honeymoon. He also lost the photos with a poorly timed crash & drive reformat. So he wants to borrow the card and recover the data. Thankfully I have not used the camera since he returned it despite thinking I should.

Luckily I ran across A Computer Repair Utility Kit You Can Run From a Thumb Drive

I didn’t like the setup of Photorec as it runs through the command line. Navigating the tree was confusing at best. It did recover 1,166 photos / 3.62GB for me.

Not trusting a single method, I also tried Recuva. That worked a little better. It reported 1,395 files found. However, 177 were unrecoverable. Getting 1,218 pictures / 3.78GB back was 52 / 160MB better than Photorec. Though many of the “recovered” pictures just say: Invalid Image. Maybe they really are Raw?

While trying to use Restoration, it crashed the first time. Not sure why. It was fine the next time, though it only found 4 photos.

Filename: Photorec doesn’t restore files with anything like the original name. Recuva and Restoration do.

Meta Data: OSes and image editors know about the EXIF data in pictures. All the Photorec pictures have date taken. Most of the Recuva pictures do. Guess I could see if only 52 pictures are missing the EXIF? That might explain why Photorec lost some of them.

All in all, it was an fun experiment. I am not curious how these stack up against of the proprietary software? Why pay $40 when these are better?

I was sent a RAR file at work. I didn’t want to install WinRAR because it wasn’t clear whether educational institutions are business or not. The PowerArchiver software worked used only pretended the capability of opening the file.

A better solution might be 7-Zip? It supports RAR.

“Who is the most wired teacher at your college?” (A Wired Way to Rate Professors—and to Connect Teachers)

Although the university runs workshops on how to use Blackboard, many professors are reluctant, or too busy, to sit through training sessions. Most would prefer to ask a colleague down the hall for help, said Mr. Fritz.

Professional support is too intimidating, cold, careless. Support fixes the problems of others who created problems for themselves:

  • choices made in software to use
  • configuration choices
  • mistakes logic in processing

The concept of identifying the professors who most use the system is a good one. We already track the amount of activity per college or university in the University System of Georgia. The amount of data (think hundreds of millions of rows across several several tables)  would make singling out the professors a very long running query. Doesn’t mean it is a bad idea. Just don’t think it is something we would do with Vista 3. We probably could with Vista 8 which uses a clean database.

I’d like to see two numbers:

  1. Number of actions by the professor
  2. Number of actions by the all classes the professor teaches

Ah, well, there are lots of other reports which need to be done. Many more important than this one. 

Some questions from the article: “Will colleges begin to use technology to help them measure teaching? And should they?” At present, to create such reports, IT staff with database reporting or web server skills are needed. Alternatively, additonal applications like Blackboard Outcomes System can provide the data. The real problem is the reliability and validity of the data. Can it really be trusted to make important decisions like which programs or employees are effective.

Heh. Blackboard Vista is headed for a brick wall? Who knew?

7. Course Management Systems are Dead! Long Live Course Management Systems! Proprietary course management systems are heading for a brick wall. The combination of economic pressures combined with saturated markets and the maturing stage of the life cycle of these once innovative platforms means that 2009 may well be the year of change or a year of serious planning for change. Relatively inexpensive and feature-comparable open source alternatives combined with some now learned experience in the process of transition from closed to open systems for the inventory of repeating courses makes real change in this once bedrock of education technology a growing possibility. As product managers and management view these trend lines, I think we might see incumbent players make a valiant effort to re-invent themselves before the market drops out from underneath them. Look for the number of major campuses moving (or making serious threats to move) from closed systems to open ones to climb in the year ahead. The Year Ahead in Higher Ed Technology

It is true the big player in proprietary CMS / LMS / VLE software has lagged in innovation for quite a while. Remember though Blackboard bought WebCT and kept around the other product while hemorrhaging former WebCT employees. That alone kept them extremely busy not to lose every customer they bought. The next version, Blackboard 9 should be available soon. That is the litmus test for their future success.

Bb9 is a newer version of Academic Suite, aka Classic. There is no direct upgrade path from CE / Vista to Bb9. There is a Co-Production upgrade path where one can run both versions side-by-side with a portal interface to access either version without having to login again. Content still has to be extracted from the old and placed in the new. (Since we are running Vista 3 and Vista 8 side-by-side now, this doedsn’t give me warm fuzzies.) This was the upgrade path some WebCT and Blackboard clients took getting from Vista 3 to 4 only to find Vista 4 was junkware. Similarly, those leaving CE4 for CE6 were frustrated by the move. So, I would predict:

  1. Those on Classic 8 now will go to Blackboard 9 ASAP.
  2. Smaller colleges on CE 8 who through turnover no longer have the people burned by the CE4->CE6 migration will probably move to Blackboard 9 this summer prior to Fall.
  3. Smaller colleges on CE 8 who still remember will migrate after AP1 (maybe a year after Bb9 release).
  4. Larger colleges on CE or Vista 8 will move some time between AP1 and AP2.
  5. Consortia groups like GeorgiaVIEW, Utah State System, or Connecticut State University System will wait and see.

That last group doesn’t take change easily. They have the nimbleness of a Supertanker cargo ship.

I am still waiting for the tweets about Moodle and Sakai, the open source alternatives, to change from in general “X sucks, but at least its not Blackboard.” to “X is the best there is.” If “at least its not Blackboard” is the only thing going for the software, then people will stay where they are to see where things go. There needs to be compelling reasons to change.

Unfortunately the cries of the students and the faculty in the minority are not enough. Most people are happy enough. They can accomplish the important things. They get frustrated that IT took the system down, data center power issues, network issues, or a performance issue. None of which go away by picking FOSS.

Cohen says:

Open-source code is generally great code, not requiring much support. So open-source companies that rely on support and service alone are not long for this world. The traditional open-source business model that relies solely on support and service revenue streams is failing to meet the expectations of investors.

The whole point is to have a model producing great code. As these open source companies try to be everything to everyone, they eventually hit the same issue as proprietary companies: Bloatware. The software starts to suck and the users abandon the ship for another product which seems to do the same job better.

Do you talk about computers, software, or web sites as manifesting human-like behaviors? Personally, I have.

At work, we manage several machines who collectively provide the GeorgiaVIEW service. When a machine is completely unresponsive, then we refer to it as having died. When a machine stops a performance issue, we refer to it as happy. I call my car cranky when it fails to run well.

Besides, electronics and vehicles, are there other examples?

Be more secure! Upgrade today.

Want better functionality? Upgrade today.

Save a developer! Upgrade today.

The save a developer thing is the impetus for this post.

The upgrade today mantra annoys me.

  1. Software rarely spends enough time in alpha and beta cycles to to identify all the issues.
  2. People have been so burned by using software in alpha and beta cycles, they are hesitant to try upgrades and help determine the issues.
  3. This lack of attention to the problems ensure, versions 1.0, 2.0, n.0 typically have a ton of unknown problems or are even less secure at times.

Unfortunately, the vendor who makes the application platform we run, Blackboard, has a philosophy to look at new web browsers while they are in beta but not actually work towards fixes for the new browsers until after the products are released. With most releases of Java or supported web browsers (Internet Explorer or Mozilla Firefox), Blackboard heard the complaints by the early adopters and released within a couple months an update which resolved the reported issues.

The students and faculty members fail to understand the issue. I think I do. Blackboard (like WebCT prior) understands there are differences between beta and final. Some of us argue these differences are usually minor. However, this is all asking someone to predict the future which we know is haphazard at best.

Long alpha and beta cycles allow more users to get involved, give those back to the developers, have them fixed before the version release. Burning users with buggy software ensures their lack of faith.

In posting a comment to a friend’s WordPress blog, it came up with the error:

Error: This file cannot be used on its own.

I was responding to a comment, so I doubted that he broke his blog between making a comment and my response. So I went looking though my own install. Essentially, at a shell I used

find . -exec grep -l "This file cannot be used on its own." {} \;

to locate the file involved is wp-comments-popup.php. This file contains code which checks for the HTTP_REFERER variable has specific values equal to the path and file name for the comments page. If this is not the case, then it should throw this error. The file mentioned in the error is wp-comments.php.

Its seems that I had configured my web browser not to pass the HTTP referrer to web servers, so the check failed and threw this error.

Maybe the WordPress developer who designed this has no idea about the ability of web browsers not to send a referrer. Searching for the error on the WP site yielded nothing. From the tons of comments about people hitting this error, lots of people turn off sending referrers.

Solution for those leaving comments: If you attempt to leave a comment and see this error, then enable referrers. WordPress actually has a decent article on enabling HTTP referrers for a number of different pieces of software.

More friendly error for WP blog owners: Edit wp-comments-popup.php. Change

die (‘Error: This file cannot be used on its own.’);

to

die (‘Learn how to <a href=”http://codex.wordpress.org/Enable_Sending_Referrers”>enable HTTP referrers</a> to fix this. ‘);

So far I have either been oblivious or lucky. Some people like my pictures which could mean they are downloading them and even representing them as their own. No amount of HTML or JavaScript technology can prevent this. Even watermarks have questionable efficacy as people get better.

Google’s Picasa is my current image editor. With it, I am able to manipulate photographs easily prior to posting them online. For everything it does, Picasa does a fantastic job. One of two things* it lacks is adding a watermark. If it automatically did this at the time a photo was saved, then I would definitely be a happy user. Maybe it will hit the features of Picasa 3?

Years ago, I knew how to do add a nice watermark in seconds with Photoshop 6 and 7. Over the last hour or so I have been playing with GIMP to accomplish the same. This has been slow going. First, in GIMP 2.2.3, the software crashed each time I opened the text tool. Now that I am on 2.4.5, the text tool works. Second, I have not found anything similar to the hand tool.

I followed a GIMP watermarking tutorial for one as it was better detailed than another I attempted to follow and was frustrated at not being able to find what it told me to use.

So, I am curious…. What do you use for watermarking your images?

* The other is splicing together multiple images.

Tier1 support was contacted by a student to report a problem with a quiz. Some sort of anti-malware software complained about code in the quiz. Tier1 support replicated the issue. Their software identified it as belonging to MyWebSearch. Very bad news.

Tier2 support suggested the student remove the MyWebSearch toolbar. Tier1 escalated to Tier3 who claimed that because Tier1 and the student both saw the issue, the malicious code must have been inserted into Vista by the instructor. Tier3 also escalated it to me… Tier4.

From the email conversation, no one looked at the quiz in question despite Tier2 and Tier3 both having the access (and Tier2 being involved in instructional design).

*headdesk*

Instead of providing my own equally valueless speculation, I got a copy of the quiz and looked at the HTML. Eventually, I used WinMerge to compare the problematic quiz and a previous quiz side-by-side.

Turns out the difference between the two is the use of WebEQ Java applets. (The problem-free quiz used GIF images instead of applets.) I think the anti-malware software of the student and Tier1 both reacted to the Java applet. Here is one of the items used.

<applet code=”webeq.Main” archive=”/path/to/jar/WebEQ2Applet.jar” width=1899 height=40 align=middle><param name=eq value=”<math> <mrow> <semantics> <mrow> <mi>f</mi> <mo stretchy=’false’>(</mo> <mi>x</mi> <mo stretchy=’false’>)</mo> <mo>=</mo> <mn>2</mn> <msup> <mi>x</mi> <mn>2</mn> </msup> <mo>+</mo> <mn>3</mn> <mi>x</mi> <mo>+</mo> <mn>1</mn> </mrow> <annotation type=’MathType’/> </semantics> </mrow></math>”><param name=color value=”#ffffff”><param name=parser value=”mathml”></applet>

The way WebEQ passes the parameters as XML inside a parameter tag strikes me as very odd. Maybe causes anti-malware to trigger a false positive?

These applets have been a problem in the past, as many were hard coded into the HTML prior to Vista incorporating this into the equation editor. Vista instead, uses a parameter block to pass the values to the applet.

I think our instructional designers are moving away from use of WebEQ as it has been consumed so much of their time, they hate it now.

« Older entries § Newer entries »