I definitely agree on this. Memory usage and 'strange behaviors' are the worst of the issues. I've been using it for 4+ months and continue to see problems in constructs that follow the guidlines yet don't work. This is probably the most experimenting I've done during development for any language and I've worked with a bunch.
Overall I like the system and it seemed the best solution available in January. However, google just put out an sdk that takes java code and compiles it into dhtml. Maybe someone has looked at it, I haven't just because it is too new. It may become a competitor next year after it really works. I hope Laszlo continues to grow and solve these growing pains. Laszlo's success will be dependent on how fast this platform can solidify. Cheers, Bruce > On 6/20/06, Geert Bevin <[EMAIL PROTECTED]> wrote: > > This was the first article: > > http://rifers.org/blogs/gbevin/2005/3/8/wasting_time_with_laszo > > As someone who has been using Laszlo for about two weeks, some of your > initial reactions resonate with me. In particular, you said, "Now > this is already bad, but it gets worse since most of the times it > never works as you intended the first time off. The combination of > bugs, vaguely documented features, abstract examples and crippled > features (like the stripped down version of Javascript and very basic > XPath support) forces you often to try out everything three or four > times before you get the behaviour you want." > > I wouldn't word it quite as strongly. In particular, I think that > most of the documentation is quite good. Still, I would agree that my > biggest gripe is also that "it never works as you intended the first > time off". > > First, XML is an inherently unwieldy and verbose syntax. I always > have to deal with a slew of syntax errors before the thing will even > compile (much more so than other programming languages I have worked > with). > > Second, parts of the API lack consistency, so I'm constantly looking > things up. For example, in some parts of the API, you refer to other > methods by using delegates, in other parts you simply use dot > notation, and in others, you pass two arguments -- the object and then > the property name as a string. Using delegates all the time would be > more uniform, but because delegates need to be explicitly > memory-managed, I dread using delegates. > > Next, Laszlo is designed so that there are at least three ways in > principle to accomplish every task (in tags, in scripting, or some > combination). In practice, however, only one way actually works for a > given problem and it takes trial and error to figure out which way to > do it. Since the tools are so limited for tracking down problems, it > is a rather slow process to isolate the code which is causing > problems, and try every variation to get it to work. Many times, the > problem has to do with the fact that it is difficult to predict the > order in which your code will be inited and executed. > > Finally, because the coding portion of Laszlo is based on Javascript, > which I consider to be a rather "weak" language, expressing complex > logic and verifying its correctness is difficult. > > That said, I still think Laszlo is the best option I've encountered > yet for deploying snazzy-looking applications with no install over the > web. The look and feel of the built-in animating user-interface > components are excellent, and it is relatively easy to create your own > custom UI elements with the task in mind. Event handling features and > constraints simplify many typical ui challenges. > > --Mark > _______________________________________________ > Laszlo-user mailing list > [email protected] > http://www.openlaszlo.org/mailman/listinfo/laszlo-user > > _______________________________________________ Laszlo-user mailing list [email protected] http://www.openlaszlo.org/mailman/listinfo/laszlo-user
