Comments below.... -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Monday, January 10, 2005 10:19 AM To: jug-discussion@tucson-jug.org Subject: Re: [jug-discussion] JDJ: Featured article on Spring
Could you elaborate more one item number 3? Come to the jug on Tuesday, and ask again. It will be elaborated upon. In addition, how is scalability with regards to spring? ** There is no issues regarding scalability and Spring. Its transaction support is faster than most commercial J2EE containers. It is even faster if you consider that you can rely on local transaction if you are only talking to one datasource w/o the overhead of JTA when it is not needed. Spring is not a bottle neck. You can use it with J2EE app server to get scalability. Let's discuss this Tuesday as I can say a lot more than I am saying or have time to write. Does it scale as well as EJB? ** Yes in most cases. In some fairly odd cases, there is one commercial container that has a speed advantage in some edge cases. More likely Spring is the speed winner. Scalability will still be achieved with your app server not Spring. Spring can work in a J2EE app server. Spring does not replace an app server. Spring does not (yet) have clustering support. On Sun, 9 Jan 2005, Warner Onstine wrote: > Contrary to popular belief we don't all do Web development (I do, but > that's me). As for advocating the use of Spring over EJB's I do for > the following reasons: > 1) Testability - You can much more easily mock up the classes you need > and drop them into a Spring container (if need be, this isn't even > necessary if you want to test singular classes) > 2) It is a lightweight container - it doesn't require a full-blown EJB > container in order to run (which also means that I could use Spring in > the client side) > 3) Spring allows you to use your own beans and doesn't require that > you tie yourself to a container > 4) Spring is much more than persistence (in fact that isn't even it's > primary purpose, it happens that is it's current raison d'etre because > of it's built-in ORM services), it makes it relatively painless to > swap out Hibernate for JDO, iBatis, etc. > > What else? I never understood the reasoning behind EJB's (especially > the fact that I had to create 4 or 5 files just to persist something). > I will be the first to admit that I have not had a lot of experience > with EJBs, but what I'm doing currently doesn't require me to, neither > will my future development. Also, one of the key things that I love is > that Hibernate's main developer, Gavin King, is helping to craft the > new EJB 3 spec. This tells me that Hibernate is onto a good thing and > Sun realizes it. > > I will let other's (who have had much more experience with EJBs to > explain why they have chosen to move away from that, but I've heard > many a story from developers (in this group and at many conferences) > who have had absolutely horrendous experiences with EJBs (even with > XDoclet added in). > > -warner > > On Jan 9, 2005, at 6:08 PM, josh zeidner wrote: > > > > > One question: is there anyone on this list who does not advocate the > > Spring Framework? > > > > -josh > > > > --- josh zeidner <[EMAIL PROTECTED]> wrote: > > > >> > >> well Ollie, I have to hand it to you, at least your ad hominem > >> attacks are entertaining. Looks like weve got a lot of Spring > >> peoples here- sorry guys, didnt mean to burst your bubble, rock the > >> boat, upset the apple cart, etc., etc. Just looking for an > >> educated reply to some points... > >> > >> --- Ollie <[EMAIL PROTECTED]> > >> wrote: > >> > >>> Me thinks the lady protesteth too much. > >>> -----Original Message----- > >>> From: josh zeidner <[EMAIL PROTECTED]> > >>> Date: Sun, 9 Jan 2005 14:52:43 > >>> To:jug-discussion@tucson-jug.org > >>> Subject: Re: [jug-discussion] JDJ: Featured > >> article > >>> on Spring > >>> > >>> > >>> Nicholas, > >>> > >>> Of course Spring is simpler! because it doesn't try to solve as > >>> many problems, instead the developer / architect has to write and > >>> support that code him > >> or > >>> herself! At the cost of the customer! > >>> > >>> I am surely not the only one in the Java world > >> who > >>> is suggesting these points. I'm not baiting > >> anyone( > >>> what do you think I stand to gain? ). Ive already been in enough > >>> heated discussions recently( maybe some of you caught the > >>> 'municipal Wifi' thread ). This is a list where Java subjects are > >>> discussed( I think > >> ). > >>> > >>> > >>> In the end I would hope that a > >>> consultant/developer > >>> is offering his or her client either a competitive advantage or a > >>> cost savings. In my experience, > >> the > >>> 'architects' rarely do either. > >>> > >>> I don't see it as some kind of coincidence that > >> as > >>> soon as EJB development became affordable and accessible, that > >>> some 'expert' declares it > >> unusable > >>> for some reason, and invents some new technology that will save > >>> the day( and simultaneously a million professionals crop up ). If > >>> OSS is what you want, you can do that with Tomcat/JBoss. I can > >>> find > >> qualified > >>> developers through Sun's program, why delve into local > >>> meritocracy/salary rating scheme if I know 1) what this technology > >>> does, 2)what people can utilize > >> it, > >>> 3)what platforms are reliable. EJB is much > >> cheaper > >>> to > >>> use these days- and the technology hasn't changed. > >> > >>> > >>> Most of these Spring architects were selling EJB 3 years ago when > >>> it payed well from the developer > >> end. > >>> > >>> What are they telling those customers today? > >>> > >>> These aren't bait questions, I am actually > >> looking > >>> for an answer to justify Spring. > >>> > >>>> Those who have used both (e.g. me, e.g. Rick) > >> know > >>>> which one we prefer, > >>>> AND which one is simpler. > >>> > >>> it certainly is better from the developer perspective. But if I > >>> were a project manager, and > >> I > >>> am interested purely in the costs of development > >> and > >>> support, do you really think that Spring is going > >> to > >>> be a better solution? What, if anything justifies me adopting a > >>> panoply of object databases, messaging frameworks, and object APIs > >>> instead of an OSS platform that has wide industry latitude and is > >>> very cost effective? It just does not really make sense for the > >>> project manager. > >>> > >>> As far as the UML, Grady Booch world is > >> concerned, > >>> I > >>> have seen millions wasted on UML. UML is both > >> more > >>> complex and less discrete than most programming languages. UML is > >>> another technology that has had its chance during the CORBA era, > >>> and failed to offer > >> any > >>> serious value. If you know anything about the history of latin, > >>> by the end of the roman empire it had become so complex and > >>> formalized that the only people who were qualified to write legal > >>> statutes had to > >> train > >>> for 20 years. The Latin that we know from this period was not > >>> even spoken- much the same way that UML is often spoken about, but > >>> rarely used in practice. > >> It > >>> was the downfall of Rome, the costs of running the empire were too > >>> high to justify its existence. In come the barbarians. Usually > >>> when I comment on > >> this > >>> someone immediately produces a UML document, but being in the > >>> trenches I can tell you that there is way > >> too > >>> much overhead involved in actually utilizing this visual language. > >>> > >>> If I were a manager I would be very wary of Spring. > >>> Sure there is less there, but that doesn't mean > >> that > >>> it is going to cost less. IBM has a lot to do > >> with > >>> this. They are a once the main source of industry research data > >>> and the primary productizers. If > >> that > >>> doesn't spell *inflated costs* to you- take a look at the > >>> pharmaceutical industry. Microsoft has a different character but > >>> they are equally troubled > >> by > >>> internal bureaucrats. I can tell you that this combination > >>> coupled with what looks like a fallout from Sun will create a lot > >>> of turmoil. Other > >> kinds > >>> of > >>> apps will take center stage... > >>> > >>> > >>> -josh > >>> > >>> --- Nicholas Lesiecki <[EMAIL PROTECTED]> > >> wrote: > >>> > >>>> Not taking the bait. You shouldn't either Rick. > >> I > >>>> don't want 40 emails > >>>> in my box about EJB vs. Spring. Take it up on BileBlog or the TSS > >>>> forums or somewhere if you must. > >>>> > >>>> Those who have used both (e.g. me, e.g. Rick) > >> know > >>>> which one we prefer, > >>>> AND which one is simpler. Those interested in debating the merits > >>>> of spring here can do so without resorting to > >> phrases > >>>> like: > >>>> > >>>> "egghead $100K architects" > >>>> > >>>> and > >>>> > >>>> "seeking the blessing of high priest Grady > >> Booch" > >>>> > >>>> yeesh! > >>>> > >>>> Nicholas Lesiecki > >>>> Software Craftsman, specializing in J2EE, Agile Methods, and > >>>> aspect-oriented programming > >>>> m: 520 591-1849 > >>>> > >>>> Books: > >>>> * Mastering AspectJ: http://tinyurl.com/66vf > >>>> * Java Tools for Extreme Programming: > >>>> http://tinyurl.com/66vt > >> > > === message truncated === > > > > > > > > > > __________________________________ > > Do you Yahoo!? > > Yahoo! Mail - Find what you need with new enhanced search. > > http://info.mail.yahoo.com/mail_250 > > > > -------------------------------------------------------------------- > > - To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
