In the scope of our research project, we're studying software upgrade problems and looking for testimonies; we are also going to hold a lottery to reward submitters of useful testimonies.
We're looking for information on software upgrades that result in faulty or unexpected behaviors. More precisely, we are looking for instances of the following problems:
- Upgrades that do not integrate correctly into the system (perhaps because of linkage problems or other incompatibilities) or those that cause one or more applications to crash after the upgrade.
- Upgrades that produce different results depending on the configuration or environment in which they are installed (i.e., they work for some users but not for others).
- Upgrades that affect the behavior of other applications (i.e., incompatibility, broken dependency, different outputs, etc).
- Packaging problems, especially those that aren't simple package conflicts
- And of course, upgrades that are simply buggy.
If you've had similar problems in the past, we would highly appreciate if you could reply to us and describe your experience with as much details as possible.
To be useful, your testimony should include the following elements:
- A detailed description of the problem.
- The name and version number of the software being upgraded, the operating system, distribution, kernel, and all applications involved.
- If possible, an explanation on how to reproduce the problem and a link to websites or forums discussing the issue.
In order to recognize your effort in providing the testimony, we will hold a lottery to select four winners who will each receive a $50 (50 american dollars) amazon.com gift certificate.
All useful testimonies will participate in the lottery. People providing multiple useful testimonies will participate in the lottery once for each testimony. The deadline to send testimonies is January 19th, 2007. Winners of the lottery will be informed shortly after.
Please direct all testimonies to email@example.com
Please note that our research project is a joint effort from two different laboratories at EPFL (LABOS and NSL), Switzerland, and the DARKLAB at Rutgers University. We are not affiliated with, nor sponsored by any company. The results of this study will be used to evaluate a new infrastructure aimed at improving software deployment and testing.
We thank you very much in advance for your participation.
With best regards,
School of Computer and Communication Sciences, EPFL