Level 2: QCD Perspective
------------------------
(B. May for the QCD group)

+ Overall QCD comments were positive with some suggestions for improvement

Physics View:
  + It worked well, we got lots of data
  - BUT Had to throw out lots of data due to inefficiencies:
     - L1 jet seeds very inefficient for finding jets
     - L2 <> Reco jet algorithm, another inefficiency
  * Suggest making L2 as close to RECO as possible!
  * We would like a .NOT. option for L2 to VETO on objects

Comments on Testing/Releasing:
   - Some released unfixed previous released fixes
   - Testing on different platforms troublesome
   - Documentation of "emergency" fixes scarce
   * Testing of code most important!
   * Resist rush fixes! (1-2 days doesn't matter)
   * Document releases/fixes thorougly and make easily available to public!
   - Some cuts made in code that were not available as tool parameters and
     thus hard to track changes in value with new code releases
   * Enforce all "cuts" be changable via paramter list?
   * Archive all new code with associated L2 exe version number?

Simulator:
   - Some people thought simulator clumsy and full of jargon
   * Would like possibility of linking to front end of RECO/analysis programs

• The VETO option was proposed by Rich Astur at the beginning of Run 1b, but dropped due to lack of interest
• Strongly support the ``1 or 2 days don't matter'' comment
• The request for more tool parameters is opposed by triggermeister comments
• All code is indeed archived in CMS, and accessible by version number
• All changes were documented, but agree that it was not summarized so as to be easily used