Fork me on GitHub

Little bit confuse - Why 0.760 ?, when 0.8 ? what happened  Bottom

Go to page 1 - 2 [+1]:

  • I fell little bit confuse – we have to many version of postnuke.: 0.726 , 0.726+XTE, 0.760 ( in progress). Don’t take it like a attack – I am very happy to hear that we will have new version of my favorite system but…

    From management and support point of view I fell that new version will generate a little mess
    Why ? :
    - because we will have three active version of the system to support and fixing.
    - because we change road map of the system to often
    - because many ppl are still not convince to 0.750
    - because for many ppl upgrade is very serious task and they will not risk to do it to often

    And finally as I wrote in comment to latest information about postnuke security advisor – we can’t leave oldest version without support ( with all respect to authors but to we cant just say that all must upgrade to latest version to have system safe – even MS support oldest version of own system)
  • The rationale behind things is that once complete .726 -> .8x would have been a big leap with virutally 90-95% of the codebase replaced or rewritten. This, IMO, was far too much. You say people are not convinced about .750 - this contained only about 5-10% of the necessary code changes so you can see that the move straight from .726-.8x would have only made this worse.

    The main issues were that I didn't want .8x being seen as incompatable with previous versions nor buggy. Moving directory from .726-.8x would have meant both of those. The latter due to the amount of new code that would have come online in that release.

    So we had .750. This got the core just about right (making the use of the DB more efficient, tidying the structure etc.). It also introduced the templating environment. I think the amount of pnRender based modules shows that the developer community is certaining 'convinced' by .750. I know, personally, that I could never work with pnHTML again.

    .760 came about initially out of meeting some personal requirements. We then felt that a fair number of modules were of a releasable state. So it made no sense to release older versions of the, say, the modules module when the updated version was there and working. So .760 contains bug fixes, enhanced template environment (xanthia is much improved) and further code from .8x.

    With regard to upgrades the rule for any open source project is to use the latest stable version. Even if we had the resources of Microsoft to maintain and support older versions (which they don't) I wouldn't want thier track record. With .760 being very close to release .726 will be two stable versions out of date very soon.

    With regard to the roadmap the goals are still the same and progress is being made towards these; in fact .760 is an active demonstration of the goals and progress - each updated or new module comes directly from the .8x codebase. Consider the .750+ releases to be pitstops along the way to take stock of what we have. Releasing complete modules from .8x now gets them tested by a wider audience, validates a development techniques and reduces the number of initial problems for .80 itself.

    -Mark

    --
    Visit My homepage and Zikula themes.
  • I cant agree with your roadmap ( for me it’s a clear way for near future without release unexpected version like 0.760) - if you release next version like 0.770 0.780 etc you will see what’s happed with community.

    I understand the technical aspects and advantage from new version but - ppl have the old one and we (or I and my team in Poland ) should help them in upgrading process or in keeping system safe even without upgrade.

    I don’t really know how many afford is needed to prepare fix for 0.726 but if its possible please prepare it.
  • And then what about .723, 722 etc. - where does it stop. When each of the three security releases for .726 came out (resulting in .726-1, .726-2 etc.) this was the current stable release - no patches were provided for .723 or earlier at that point. This has always been the case - patches are supplied for the latest stable release.

    There will be no offical .726 patch but that doesn't stop anyone from creating an unofficial patch and supporting it themselves.

    For the .760 release ultimately the team felt that there was enough new code in there to warrant a release. This, IMO, is a perfectly valid release strategy. I'd rather the new modules were out there now being tested that sitting in CVS not being used.

    As i've said, IMO, going straight from .726-.80 would have too big a single leap; with too much new code and, therefore, the potential for too many bugs generated by not so widely tested code.

    -Mark

    --
    Visit My homepage and Zikula themes.
  • hazan

    I cant agree with your roadmap ( for me it’s a clear way for near future without release unexpected version like 0.760) - if you release next version like 0.770 0.780 etc you will see what’s happed with community.

    ..as Mark said, replacing 80-90% of the code base in a single upgrade would be far more detrimental than incremental releases.

    hazan


    I understand the technical aspects and advantage from new version but - ppl have the old one and we (or I and my team in Poland ) should help them in upgrading process or in keeping system safe even without upgrade.

    ..you can check the files for differences and apply the patches as necessary.. this has also been stated, by Andreas, I believe.

    hazan


    I don’t really know how many afford is needed to prepare fix for 0.726 but if its possible please prepare it.

    ..should the fixes be backported to v.711 as well? ..the .750 release was a giant security enhancement in itself, but did not offer substantial features over .726.. consequently .750 could be considered a security update and a highly recommended one at that. if you choose to use outdated software, you do so at your own risk. It's not in the interest of time or resources to attempt to backport security fixes/enhancements pre .750.. which again contained very significant security enhancements throughout the core.

    ..I do applaud your willingness to support .726 and older, but you might want to impress upon your users the importance of upgrading to "at least" .750.

    cheers :)
  • ..eek.. Mark beat me to it.. it's a little redundant now ;)
  • Quote

    we (or I and my team in Poland ) should help them in upgrading process or in keeping system safe even without upgrade.

    help them in upgrading - that's the only way for a secure system - and this is not only valid for postnuke but also e.g. for PHP itself.

    Quote

    I don’t really know how many afford is needed to prepare fix for 0.726 but if its possible please prepare it.

    since it's only about a couple of files it's not that big effort - but due to various other issues merging those fixes into .72x doesn't make it *really* secure and therefore there won't be an 'official' fixpackage for outdated versions. as invalidresponse already mentioned postnuke .75x was in some way the 'fixpackage' for .72x and it's highly recommended to use the latest stable release.

    --
    regards from germany
    ..::[Zikula Application Framework]::.. ..::[SEO-Blog]::.. ..::[CMS Sicherheit]::..
  • InvalidResponse

    ..should the fixes be backported to v.711 as well?


    <sarcasm>you mean the PN team WON'T be supplying a fix for .711? what about PHP-Nuke? shouldn't they back-port it all the way back to Thatware!</sarcasm>

    8) :shock:
  • THe only mistake that has been made was made before 750 was on the radar, and may have even been with previous teams, that of saying that "7.?? is the last release before .8" Think about the move from Windows 3.1 - 95 -98 - ME - XP Now think if it was 3.1 to XP Culture shock with 99% of stuff not working.

    --
    Home Page | Find on Facebook | Follow on Twitter
  • craigh

    InvalidResponse

    ..should the fixes be backported to v.711 as well?


    <sarcasm>you mean the PN team WON'T be supplying a fix for .711? what about PHP-Nuke? shouldn't they back-port it all the way back to Thatware!</sarcasm>

    8) :shock:


    What about that
    <joke><sarcasm>

    Dear users.
    Before we release version 0.8 we will release another version “0.77” which is only a fix pack but cant be issue as a fix pack to 0.760 because 80% of core code has been changed.

    We also plan to release 0.78 - all its depend on our roadmap but unfortunately we lose this damn map somewhere.

    Anyway upgrade is necessary and must be perform immediately – even before 0.78 because we will not support 0.760 which as you know is two moths old version.

    We will promise that 0.78 will be the last version before 0.8 ( if we will find this damn roadmap we will make the correction)



    And who is better? :)</sarcasm></joke>
  • Ah for cryin out loud, why don't we just stop work on PostNuke at all. If we release a new version it will cause problems for people. No one needs the new technology anyway I guess..

    --
    Home Page | Find on Facebook | Follow on Twitter
  • mhalbrook

    Ah for cryin out loud, why don't we just stop work on PostNuke at all. If we release a new version it will cause problems for people. No one needs the new technology anyway I guess..


    Come on - dont be so serious – somebody joking don’t you see that ?

    Back to this serious subject - you will not convince me but i must accept your way of upgrading this system and the reason.
  • Linux releases allot more minor upgrades then PostNuke, and I have had many problems with each one. I dont see a problem with postNuke releaseing .760

    Infact, some of the bugs in .750 should have been addressed earlier. People will upgrade if they feel they need the extra funtionality. It is silly to have years pass for each new version. I dont agree with a slow release date.
  • Hehe, well I know everyone here will be surprised to hear this, but believe it or not my wife and I have been known, from time to time, to disagree. Upon occasion rather vocally.

    That doesn't mean we don't get along rather well at other times, though.

    Disagreeing is normal.

    I must say that those of us, like me, who don't have the smarts to do this work really appreciate the effort of those who do. Sure doesn't mean I wasn't really hoping for 0.8 though. But I really enjoy and appreciate what we've got here, as well.

    Thanks guys!
  • Hazan,

    With regard to the roadmaps and the like it all well and good to right this kind of stuff down but one should never be so inflexible that things can't change. Perhaps it was a mistake to say .750 would be the last .7x release but you seem to be suggesting that people aren't allowed to make mistakes. We're all human; we all make mistakes.

    Additionally opinions evolve over time; what we think is a good idea one day won't necessarily be such a good idea later once additional information is available or knowledge gained. One should never be so inflexible as to not allow themselves a change of mind or opinion occasionally.

    You also seem to making the assumption that .750 & .760 have delayed .80. This simply isn't the case. .80 has always been an ambitious project. .80 involves a complete rewrite of every single module in package. This was always going to take time. The changes in PN's team over recent years haven't helped. The team is now far more stable and, hence, pace of development has changed resulting in two full releases in 6 months and with much other code closing in on completeness in cvs.

    .750 was .726 with a some generic security enhancements (tidying of globals handling etc.) + the .8x temlpating system and mailer module. .760 is .750 + bugfixes + many modules from .80. Basically your getting a newer, more capable PN now with features, modules etc. introduced to make everyones life easier and better. Each of these releases takes PN closer to .80 and gets a lot of code tested and validated making sure that .80 will be more stable on release.

    -Mark

    --
    Visit My homepage and Zikula themes.

Go to page 1 - 2 [+1]:

This list is based on users active over the last 60 minutes.