Letters to the Editor

By SD Times News Team

February 1, 2006 —

Following are some comments on Mike Prest’s “Runtime Issues,” a letter to the editor in the Jan. 1 issue of SD Times [page 27]:

Paragraph 1 – Mike, Java is backward compatible unless it was written specifically for Java 1.1 or Microsoft’s Java (or any version after). The problem lies with the developers, not Java. So if it doesn’t work, it is the developer’s fault. There is a UI toolkit that will work with all JVMs back to 1.1 (Nexaweb).

Paragraph 2 – Mike, seeing that you had problems/questions in the previous paragraph, it doesn’t surprise me that you have issues with stacktraces and Java application servers. Again, the problem here is not Java. CF on Java? There are bound to be issues.?Logs? Those come from the app server, not Java. Pick another vendor for your app server. If you can’t (if CF won’t let you), then get a different UI tool. You also probably should find an experienced and knowledgeable developer to help you out.

Paragraph 3 – Yes, Java has many names.?Some are useful, and some are not. Things like this happen when you have a large product, and a large company and a marketing department. Anyone who is actively involved with Java will have no problem, though.

Paragraph 4 – Mike, the reality is that computers and software development are not easy. When computers (and software) run, they are great. When they don’t, they are major pains.?As for “easy to port,” Java doesn’t need to be ported. That is one of the great things about Java—it is portable. Mind you, it doesn’t prevent you from making it not portable. Many of us develop on one OS and deploy to one or more other OSes.

Mike, Java is not perfect, but nothing is. I’ve used and still use other programming languages/platforms. They have their own pains and many times are worse.

Mark Nuttall Winston-Salem, N.C.

Defending Subversion
In his opinion piece [“Why Commercial SCM Tools Are Better Than Open-Source Tools,” Dec. 15, page 32], AccuRev CTO Damon Poole makes various false or misleading claims about Subversion, and about open-source software in general. He also implies that only proprietary SCM systems are worthy of serious consideration.

Given that his company competes directly against open-source systems such as Subversion, it’s understandable that he’d want his readers to think this, but it is a disservice to those who simply want the best solution to their change management problems.

Poole states that the Subversion project has “accomplished only the first two of their four goals, atomic transactions and fast branching.” Our goals, which number far more than four, have been listed prominently on our home page, subversion.tigris.org, since the project started. Furthermore, we accomplished the goals that we targeted for our 1.0 release in 2004, have made three new feature releases since then, and are currently working on some of the very capabilities he claims we’re ignoring. Again, this is all being done publicly and openly; on our mailing lists we regularly answer questions about features currently under development.

He also seems to underestimate the amount of commercial investment made in Subversion development. CollabNet (disclaimer: my employer) started the project in 2000, and has funded it consistently for six years, as of this writing, by employing several full-time developers, subsidizing QA and some requirements elicitation research, and providing hosting and infrastructure support. While we do not in principle agree with Poole’s claims about the link between funding and innovation, in this case he does not do justice to the amount of funding anyway.

Later, he uses dollars spent in 1994 versus 2004 to measure the increase in popularity of commercial SCM systems. This is spurious because the software industry itself has grown so much during that time; worse, it is conveniently incommensurate with growth in open-source adoption, since open-source software doesn’t have license fees. Yet we know that open-source adoption has been extremely robust during that decade. Why doesn’t Poole attempt a relative comparison, instead of uselessly comparing proprietary SCM tools against themselves?

There isn’t space here to address all of Poole’s claims, but at subversion.tigris.org/poole-response.html, we’ve posted a more detailed response, and hope that readers who saw Poole’s original piece will also see our reply.

Karl Fogel Software Collaboration Specialist, CollabNet Developer, Subversion Project

Share this link: http://www.sdtimes.com/link/29106

Letters to the Editor: Scala: Good transitional language A reader defends Scala's usability in modern development, while Andrew Binstock replies to a letter from the last issue

Letters to the Editor: Evolution of a data blob A reader takes us down Memory Lane to show us how databases came to be what they are today

Letters to the Editor: Doomed to repeat software failures A reader sees another reason why software failures still happen. Also, another reader urges caution when praising Scala

Add comment

Name*
Email*
Country

Comment
Preview

News on Monday
more>>
SharePoint Tech Report
more>>

Download Current Issue
ISSUE 3/15/2010 PDF

Need Back Issues?
DOWNLOAD HERE

Receive the print Edition?

Google Code turns 5
Google Code Turns 5, and adds a Paxos Algorithm to make the system more stable and reliable.
03/17/2010 11:16 AM EST

Test your Visual Studio 2010 know-how
Microsoft is offering free beta certification exams for Visual Studio 2010.
03/17/2010 11:08 AM EST

Microsoft lifts the hood on IE9
Microsoft is previewing IE9.
03/16/2010 01:10 PM EST