On 8 Jul., 14:08, Michael Kimsal <mgkim...@gmail.com> wrote: > I'd prefer more people spend their time making decisions about > business-level or application-level issues, rather than language or > framework issues. In the MS world, it *can* be much easier to focus > on those issues regardless of your team's experience, because almost > without exception they will be using the same tools and the same set > of libraries (framework/3rd-party/etc).
Thanks for pointing that out - you are right, in many cases there is too much discussion about language or framework to use although there might be less difference in those than in different approaches on business-level. Full ACK on this. > My limited > experience has seen that time translated into more business-focused > decisions/investigations, though it could just as easily be spent in > other techie-focused stuff instead. And yes, time can be easily wasted elsewhere too. ;-) > > In general when stepping outside the Microsoft-World you start having > > many options > > [...] > This is a benefit only to some segments, and this is a real drawback > in others. It's great that you might prefer Debian Linux, but when > the largest pool of talent for Industry X (which you are > consulting/working in) is for RHEL, your preferences shouldn't come in > to play as much as the client's preference for being able to find a > wide talent pool. Thanks for mentioning this also - I didn't think about that - I am usually focused on trying to find the optimal solution from technical point of view and not from "political" or "market dominance". But those factors may play an important role - you are right! > The 'lack of choice' in the MS world can be seen (and indeed is spun > as) a good thing in many quarters, and I tend to agree. Certainly I > *like* choice, but I also appreciate a level of standardization that > comes from intentionally leaving many choices out. Yes, choice can be a pain and be a good. My experience from development under Windows is (and this is where my mileage may vary from yours), that in various cases technical approaches ended up in a dead end and I had to find alternative ways. Our company has switched 3rd party libraries a few times because of serious problems or bad support in several cases. Having enough other options can save your soul then. Sometimes one library might work stable but is simply too slow. We are also working on applications that do use different similar components, because one thing only works well in library A and one thing only in library B although there are a lot of things that library A can do as well as library B. One might think that one of those should be enough but isn't. - There is also choice in the MS World, but usually less and that already caused me a lot of pain. - Sure, choosing between many options can also be a pain... > > OK it may take > > a while to sort out what fits better, but when you know then you get > > something that > > fits your needs and you can do a better consulting to others maybe > > having little > > different requirements. - And same applies to Java Web technologies > > (this is my current > > "battlefield" where I am still evaluating). > Go with Grails. :) I thought I should wait for Scala & sails... ;-) > I wouldn't say "often". Probably not even "usually". More effort > leads to fewer people able to learn all the ins-and-outs of every step > of the process, and leads to people taking copy/paste shortcuts or > relying on IDEs to generate swathes of non-understandable code. In my > experience, "more effort" has very little bearing on "quality". At this point my mileage varies a lot. But we should distinguish different types of effort here - e.g. effort put into analysis of business need, effort put into language and framework evaluation and so on. Too many times I have seen projects being developed without enough analysis and evaluation and that made costs exploding. > but it could also > very well be that there's benefits to the language and platform that > make it easier to provide the quality without as much effort. I think, that the language itself is just contributing to effort needed and quality. There is IMHO no doubt, that a person with long year experience using a particular language and the available self- created libraries, as well as knowing various 3rd party-libraries that integrate well, can be more productive than somebody using some new exiting stuff that offers some cool new features. On 8 Jul., 15:41, Casper Bang <casper.b...@gmail.com> wrote: > > In the MS world, it *can* be much easier to focus > > on those issues regardless of your team's experience, because almost > > without exception they will be using the same tools and the same set > > of libraries (framework/3rd-party/etc). > > [...] > Agree, this sets a lower bar for development, which some of the Java > community likes to interprets as it being a dumb, docile and > uncritical community. > The good-enough perspective is dominating on .NET, with de-facto > support for most needs leading to less-is-more; I can understand if > managers are tempted by this. Which stack is the best for your next > major product in Java? There's no definitive right answer, you can go > with something proprietary (ADF), what looks de-facto (Wicket) or you > what's official (JSF) - it feels a bit like playing the lottery. Yes, indeed - especially for me beeing a newcomer - it feels playing the lottery, because I even do not know the big pluses and the pitfalls of all those technologies. It is hard for me sort everything out. But: To me it seems, that "real" (r)evolution in software development in the last ~10 years or so is mostly coming from the Java related world (just my impression). Microsoft might have introduced some improvements in C#/.NET while copying Java, but I cannot see that amount of innovation on the Microsoft side as I see it here. There are many different approaches which I see as innovative work. I agree with you, that in Java land there might be working much more perfectionists than in the MS world. I should adopt that thinking when int comes to choosing a web technology for instance. I guess - e.g. JSF or GWT or even plain old JSP would even do it perfectly fine in most of my projects having less pain than in all my legacy projects. However, I want to take advantage of the options choosing that or those libraries/frameworks/components/technologies bringing additional benefits and help me best being efficient. That said, I know a Java developer working for years now on a big project using JSP and deploying changes manually without either using a war file and is perfectly fine with it - and I think he either doesn't look on different technologies. > Ah but then we would not have the "fun" discussions between OSGi vs. > Jigsaw, Eclipse vs. NetBeans, tabs vs. spaces, SWT vs. Swing, > Hibernate vs. JPA, Scala vs. Fan etc. etc. That's not necessarily a > stab at open source, more a realization that perhaps it's good to > leave some decisions to the experts/architects. It is important not to loose the "connection" to the final customers and user experience. In many IT projects there is a large gap between the techies and the management (on developer and customer side) and another large gap between those and the users. So even if I do not feel to be among those experts you mention, I really want to be in touch with them to ensure that technology is evolving "in the right direction". > So just as Neo found out, the real world is not as polished and pretty > - be ready for oil on your pants and cranks that don't always fit. The > Matrix had one architect, the real world don't. What I would love to see is some good overview of available technologies (not only web-technologies) with recommendations for newcomers, what they should learn first. In one of the last podcasts JSF was mentioned to be a good start for learning web development - not sure if most others have this opinion too. I also heard a lot of positive words about servlets. I think it must be possible, to offer some "good-enough" quick solution for those who do not want to bother with months of evaluation what frameworks to use. When I have particular needs that I want to take a more detailed look on the options I can do it anyway (after collected already some experience with the "standard path"). - I would like to see a podcast which focus on this issue. - At least here in Austria Microsoft is offering .NET courses for the price of a banana. I guess they are very successful here to get more and more .NET developers that way. This will certainly make their community grow fast. On the other hand it is more difficult today for one to learn Java. At least the available (online) documentation is far, far better than what I have seen for .NET so far. On 9 Jul., 01:55, Peter Becker <peter.becker...@gmail.com> wrote: > Usually > questions like "do we use Wicket or JSF as web-framework?" has an answer > along the lines of "use our internal, Struts-based one". "You want to > use an OSS library? Only if it is in the list of libraries we allow (or > you make a bloody good business case so we put your project into the > status of a pilot for the library)." > [...], but the Java space not only gives these organizations the > freedom to adapt more, it also means that all these guys fiddling with > the new toys might actually produce something that might be worthy as > part of the new standards in a few years time. Yes, while this makes sense in big companies on one hand, it also holds back innovative new approaches on the other... > The MS world doesn't seem > to have this bottom-up approach as lively. It is there to some extent, > but Java's strength is that there is so much happening that you may or > may not follow. If you are a large enough organization, then you can > easily afford some staff to watch the space and try to figure out which > new technologies suit you and when. I see a strong focus of the Java community on the real big companies, what about the large amount of smaller ones? In my particular case, I am going to put more effort now into evaluating different libraries and frameworks what fit best for the major part of my future projects (which of course is also somehow a look into the glass sphere) and then I will go with that for the most projects. Fortunately for our customers the most important thing is that the software is working and reliable and they do not ask that much what's used in there. They might be happy if it's Java, because it's a well known thing and not something that only a few guys in the world know. Our customers usually do not have enough technical knowledge to write policies about particular frameworks that should be used or should be avoided. So I have a certain amount of freedom to choose the used technologies according to what I think is the best. And this is my responsibility - choosing "good" technologies... --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to javaposse@googlegroups.com To unsubscribe from this group, send email to javaposse+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/javaposse?hl=en -~----------~----~----~----~------~----~------~--~---
