The team haven't said something because perhaps the team can understand: "Please try to avoid a quick answer and take your time to "digest" the matter."
btw, Stephen Bohlen, who probably has a better biorhythm, gave his point of view giving us another possibility that translated in work to do means: NH can stay on .NET3.5 and we can implements/deploy a ICollectionTypeFactory for .NET4 with the IPersistentCollection for System.Collection.Generics.ISet On Mon, Oct 18, 2010 at 1:17 PM, Frans Bouma <[email protected]> wrote: > > I'm already doing (doing mean exactly doing) some of those works where I > > think is the right place. > > - audit : talking with Simon and Roger and helping with some code > > - validation: working in NHibernate.Validator that has very few few > things > > in common with Hibernate.validator > > - INotifyPropertyChange writing a post to demostrate how do the > integration > > having a little performance advantage in less then 50 code lines. I'm > > ignoring any other possible support. > > - security: I have asked explication and I'm still waiting > > and I've answered. that you won't listen to what I've to say, > that's > your problem really... > > to recap: security at the entity and field level, e.g. who can read > / write property X, who can read / write entity INSTANCE Y etc. this can be > done at the db level (rhino security?) but also at the property level in > memory with intercepting code which consults an 'authorizer' which is > injected into the entity. I don't know if that last bit exists for > NHibernate, if it does, it needs to emit code into the proxies, hence my > remark about it being something one would want to use WITH Nhibernate. > > You keep on talking about NHibernate core, but that's not the point > of my argument: I'm arguing that the O/R mapping problem has been solved a > long time ago and when someone wants to use 'NHibernate' the whole scope of > features EXPECTED to be covered with 'using NHibernate' is bigger than the > core O/R mapping problem, i.e. NHibernate dll + contrib stuff + all kinds > of > other features scattered all over the place in 3rd party libs. > > > About fork NH: I have invited people to do it more than one time, after > > NH3.x we will move to Hg or Git and who want can do it even more easy. > > IF I fork it, it won't be a private repository nor will it be aimed > at 'just an nh fork like many others'. But we'll see... > > > To resume: > > I still waiting the definition of > > - NH "look like depends on the direction where the wind comes from" > > Not sure where this out-of-context remark is coming from, but as > you > don't answer/reply to my questions/remarks as well, why should I reply to a > remark which isn't 1) in context and 2) it's not even clear who said it and > when. > > > - "professional toolkit" and why NH can't be defined a "professional > > toolkit" already > > I gave you 'a' definition, do I have to rehash myself every time ? > > Anyway, AGAIN: you asked for input, I gave some input. It's up to > YOU to do something with it. However instead of constructively use > knowledge > from the O/R mapper field around you, you block everything out, and more > importantly, your fellow team members don't pick it up either. > > Now, why don't you tunnel the energy you've invested in picking > apart my reply instead towards your teammembers who haven't come up with a > list of ideas for 'vNext' ? > > FB > > > - "security" what exactly mean for NH > > > > -- > > Fabio Maulo > > > > > > El 18/10/2010, a las 10:59, Frans Bouma <[email protected]> escribió: > > > > >> On Mon, Oct 18, 2010 at 3:55 AM, Petrov Peter > > >> <[email protected]> > > >> wrote: > > >> I think Frans Bouma has unbelievable patience. We aren't far from > > > the > > >> time when Frans will say "enough is enough" and I fear it will soon > > >> stop following this dev group. It will be a loss for the NHibernate > > project. > > > With > > >> his attitude Fabio will force back other people too. For mr. Maulo > > >> it's > > > time > > >> to reconsider his attitude toward others and his sharp tone. > > >> > > >> Perhaps you can add some opinion about the initial post of the thread > > >> instead try to be "the paladin of justice". > > >> Frans can say that he want see auditing, validation, > > > INotifyPropertyChanged > > >> and security inside NH-core and I can write my opinion and ask for > > >> clarifications, then the decision is not mine but of the team. > > > > > > I wrote on Thursday: > > > "The examples I mentioned are not things you expect on the core O/R > > > mapper level, however they're all services which build on top of the > > > O/R mapper core and utilize it, which also means that if you use > > > NHibernate, and you want to use the features I mentioned, you need to > > > implement them somehow so they utilize NHibernate." > > > > > > Google groups link: > > > http://groups.google.com/group/nhibernate-development/msg/e91d2b8e8bd2 > > > 02bd?h > > > l=en > > > > > > I never said it should be part of the core, as those aren't core NH > > > services. You have apparently no clue whatsoever I was talking about, > > > which explains your replies. > > > > > > Anyway, you have a narrow focus on where NH should go to, but you > > > don't realize that this focus only targets core O/R mapper > > > functionality, a problem solved years ago already, which means that > > > there's little to nothing to do in the future except re-inventing > > > wheels, like we see already with yet another xml / code based mapping > > system and yet another query system. > > > > > > IF I'll eventually say "Enough is enough", I'll fork NH and do what > I > > > explained here. But that takes time and I've to look into what it > > > brings me, compared to the effort to invest. > > > > > > FB > > > > > -- Fabio Maulo
