|
Sponsored Links
Resources
.NET Research Library
Get .NET related white papers, case studies and webcasts
|
News
News
News
|
Messages: 37
Messages: 37
Messages: 37
Printer friendly
Printer friendly
Printer friendly
Post reply
Post reply
Post reply
XML
XML
XML
|
 |
Humor: Top 10 Things To Be Thankful for in .NET
While digesting large sums of turkey this past weekend, I woke up from my tryptophane induced coma long enough to look at all of the things we .NET developers have to be thankful for. Here's my rather tongue-in-cheek list of the technical bounty that is .NET.
Top 10 Things To Be Thankful For in .NET
Feel free to add any precious little moments of .NET bliss you care to share as well.
|
|
|
Message #147799
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Hillarious
Hey Paul,
Good job, man !!, I like C#, java without the silly coffee
references and others...
OO using VB.net is hillarious too..
I would like to add to java comment, C# ...is java without coffee, and 'core'...everywhere we go CORE JAVA, CORE JDBC, CORE Sample, if everything is surrounding CORE?, perhaps thats why we need to go to fifty sites, and download 'open source', and can plug in to get the job done. Its like I want a car, i need to run for tires, for engine, for doors, finally i have a CAR which doesnot run !!...WHAT TO DO!!
RUN AGAIN !!
|
|
|
Message #147801
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
that's like
baptist saying, "catholics aren't christians" because "it's by faith alone".
just about everything gets tangled up in religious zealotry. oh wait, it was suppose to be a joke. too bad that is the most annoying part of both java and .NET communities.
|
|
|
Message #147807
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
that's like
baptist saying, "catholics aren't christians" because "it's by faith alone".just about everything gets tangled up in religious zealotry. oh wait, it was suppose to be a joke. too bad that is the most annoying part of both java and .NET communities.
By orders of magnitude, this exist more so in the java world. Whenever you see a topic on tss.net and tss.com that favors .Net in any way. The tss.com guys come from everywhere making broad baisless statements about .Net and .Net developers. But on tss.net hardly a whisper even when it's the reverse.
Another reason why I would never want to work in a java shop, bond to be a zealot in the group who will get on my nerves.
|
|
|
Message #147809
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
that's like
baptist saying, "catholics aren't christians" because "it's by faith alone".just about everything gets tangled up in religious zealotry. oh wait, it was suppose to be a joke. too bad that is the most annoying part of both java and .NET communities. By orders of magnitude, this exist more so in the java world. Whenever you see a topic on tss.net and tss.com that favors .Net in any way. The tss.com guys come from everywhere making broad baisless statements about .Net and .Net developers. But on tss.net hardly a whisper even when it's the reverse. Another reason why I would never want to work in a java shop, bond to be a zealot in the group who will get on my nerves.
typo (baisless -> baseless)
|
|
|
Message #147901
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
wow, that makes a lot of sense
I usually have fun with it and play devil's advocate. I've been known to make rash statements in the past either for fun, or due to a lack of clear thinking (most the later).
I think perhaps it's larger because there are more java developers on community sites like TSS.NET, since well there isn't a Microsoft like entity to provide a central meeting place. The .NET developers I know only 2 of them look outside MS resources for information. The rest happily eat what MS sells. Often to the detriment of the project because they're given requirements that existing products do not provide. I saw this happen a half dozen occasions, over the last 2 years. I remember reading a blog by one of the .NET developers about Microsoft's struggle to write their own unit test and build tools. I'm too lazy to find the link right now. Probably the worse kind of religion is corporate religion. then again, that my opinion and not fact or truth. truth doesn't really exist anyway.
|
|
|
Message #147913
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
wow, that makes a lot of sense
I would agree with Peter's assessment that many .NET developers don't look outside Microsoft for guidance, and that's just sad. Microsoft as a company employees more of the truly brightest people than any company in the world. But there are literally hundreds of others who are active in the community who don't work for Microsoft.
<plug type="shameless">
That is why it's so important to have a site like TSS.NET where any developer with a good idea or something important to say can have a way to share it with the rest of the world. TSS.COM not only highlighted Java superstars, it also helped create many of them. We have that same ability to do that in the .NET community right here.
</plug>
|
|
|
Message #147932
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
wow, that makes a lot of sense
I think it's just the way Microsoft has worked in the developer community since COM. Microsoft seems to follow this pattern:
First they release the platform, provide some initial shallow guidence for the general developer public.
Second they start to get feedback from MCS and the MVP's then they start to publish guidance form real world experience.
The Microsoft patterns/practices and architecture groups have really improved the quality of their guidance.
There is still room for improvement, but I think they are on the right path.
The .NET developers I know only 2 of them look outside MS resources for information. The rest happily eat what MS sells. Often to the detriment of the project because they're given requirements that existing products do not provide.
I would say the smae is true on the java side, how many java developers you know swallow everything IBM WebSphere BEA or Oracle has to offer without considering other lightwieght frameworks or app servers that could do the job.
This is a issue on both sides, you also have to keep in mind the java community has a how many year head start and how much industry support? You don't look at 3-year child and expect it to have mastered its surroundings or cognitive reasoning capabilities.
Com'on java guys back up and let the babies grow, but your critisism is helpful even though harmful sometimes.
|
|
|
Message #147936
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
my experience is different
I would say the smae is true on the java side, how many java developers you know swallow everything IBM WebSphere BEA or Oracle has to offer without considering other lightwieght frameworks or app servers that could do the job. This is a issue on both sides, you also have to keep in mind the java community has a how many year head start and how much industry support? You don't look at 3-year child and expect it to have mastered its surroundings or cognitive reasoning capabilities.Com'on java guys back up and let the babies grow, but your critisism is helpful even though harmful sometimes.
most of the developers I know first hand do not. then again I attribute that to luck. I've worked with a few who tried to take the easy way out, but usually more than half the team calls the developer on it and demands proof.
my experience doesn't represent the market in general by any measure. The most recent experience I had with .NET the developers rarely consulted other resources. Even when I took the time to find articles from magazines, ACM and other third party sources, the suggestions were ignored. Not to over generalize, but my first hand experience, the excuse given was this, "I'm not going to use something microsoft does not recommend, and I don't have time to read the articles." That's pretty close to verbatum, minus the nasty snears I got for pointing out the obvious.
Eventually word got to manage through other individuals what was going on, but by then the management was furious at the lack of quality or thoroughness. Other's experience may be different, but I could just be very lucky and have mostly worked with good java developers.
|
|
|
Message #147939
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
wow, that makes a lot of sense
I would agree with Peter's assessment that many .NET developers don't look outside Microsoft for guidance, and that's just sad. Microsoft as a company employees more of the truly brightest people than any company in the world. But there are literally hundreds of others who are active in the community who don't work for Microsoft.
Microsoft needs to learn more about community, MSDN has been a great success for them, but one thing I like here that I don't have at MSDN, is the ability to converse or debate with you guys.
Microsoft publishes an article, I get to rank it and post a comment that other developers can't read and learn from, WTF!!!.
I can read blogs about MS employees personal shit and some technical content that I naturally feel I have to splice out the spin because they are a MS employees (WTF!!) even if there isn't any spin, though there are some exceptions, Chris Brumme http://blogs.msdn.com/cbrumme, to name one.
Microsoft is still largely a dictatorship in terms of community, but they are getting a little better at least around the VS 2005 release.
|
|
|
Message #147948
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
wow, that makes a lot of sense
I would agree with Peter's assessment that many .NET developers don't look outside Microsoft for guidance, and that's just sad. Microsoft as a company employees more of the truly brightest people than any company in the world. But there are literally hundreds of others who are active in the community who don't work for Microsoft. Microsoft needs to learn more about community, MSDN has been a great success for them, but one thing I like here that I don't have at MSDN, is the ability to converse or debate with you guys. Microsoft publishes an article, I get to rank it and post a comment that other developers can't read and learn from, WTF!!!. I can read blogs about MS employees personal shit and some technical content that I naturally feel I have to splice out the spin because they are a MS employees (WTF!!) even if there isn't any spin, though there are some exceptions, Chris Brumme http://blogs.msdn.com/cbrumme, to name one.Microsoft is still largely a dictatorship in terms of community, but they are getting a little better at least around the VS 2005 release.
Call me crazy, but what I like most about a community detached from the company is that I am free to contribute however I want. If I'm being a total idiot and misunderstanding a piece of technology, I'd rather have someone tell me bluntly, "you're being a moron." Hopefully, someone else making the same mistake using the same bad argument will see the discussion and learn from it.
I don't know how others learn, but I mostly learn from mistakes. Of course I can also post the same kinds of questions on technet, but more likely than not it gets ignored. Atleast from past experience, those kinds of posts are ignored on technet. Other people's experience may be different.
Public forums are good because people are free to scream and say non-sense. What is annoying is when an individual refuses to be logical or rational after given ample concrete proof. I'm definitely guilty of doing that in the past, but I usually come to my senses and apologize. Humans are emotional, so it's not possible or fun to stay completely objective. I also never contradict myself, well half the time. on second thought, make that sometimes.
|
|
|
Message #147949
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
0 Carbs and 1/3rd Less Dictatorship? :-)
I think Microsoft is definitely heading in the right direction about community and openness. As proof I'll point you to this article. I can tell you that I'm also very involved with INETA (International .NET Association www.ineta.org) and have seen first hand how important Microsoft considers the .NET developer community. They have also lended more than a small amount of support to us here at TSS.NET.the excuse given was this, "I'm not going to use something microsoft does not recommend, and I don't have time to read the articles." I consider developers who think like this to be the weakest link on the developer food chain. The key for me as an architect is to motivate even these people to want to be better developers, or (and this may sound harsh) to get rid of them. That is why delivering good guidance is so important for Microsoft. For those who will only listen to what Microsoft says, it's important for Microsoft to say the right things. Take a look at the SqlDataSource debate for more on this topic.
|
|
|
Message #147954
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
0 Carbs and 1/3rd Less Dictatorship? :-)
the excuse given was this, "I'm not going to use something microsoft does not recommend, and I don't have time to read the articles." I consider developers who think like this to be the weakest link on the developer food chain. The key for me as an architect is to motivate even these people to want to be better developers, or (and this may sound harsh) to get rid of them. That is why delivering good guidance is so important for Microsoft. For those who will only listen to what Microsoft says, it's important for Microsoft to say the right things. Take a look at the SqlDataSource debate for more on this topic.
the funny thing is, as a consultant I have no power to fire anyone. instead I have to just put up with it and try to lead by example. that way, atleast I have my NUnit tests, design docs, sequence diagrams, flow charts, code comments, and tutorials written and organized neatly. even if the project fails, I can atleast walk away with the satisfaction that I delivered the highest quality under the given circumstances.
|
|
|
Message #147957
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Why this site is important
This is taken from a response I made over on TSS.com during the thread about TMC going away. I think it speaks to some of the comments here about reaching beyond the MS-controlled message:
Why do you still keep TSS.net running? I doubt it will ever gain traction as .NET developers are not going to look beyond the MSDN site and maybe a couple of other ones officially sponsored by MSFT.
Couldn't agree with Floyd more - the "ServerSide format" as I like to call it is something desperately needed in the .NET community to break developers away from that purely MS-guided approach to using the technology.
IMO, it allows us to put content out that elevates the thinking of developers and architects when approaching solutions. It's true that MSDN has raised its game with its Architecture and Developer centers, and some of the employee blogs really lead the way in terms of seeing inside the company, but there's always that innate distrust when reading things from MS (no matter the quality of the info) as to whether or not you're reading something that's good for you in your job or just simply touts the party line.
Plus, in the .NET land, there just aren't that much materials to cover besides what MSDN already covers - no open source frameworks (well, very few), no competing vendors and products (no OR mapping, no EJB's, no MVC, no IoC), no religious fights, no gossips, except every some is guessing the exact time when Longhorn will come out.
There are always competing products in the .NET space to consider - databases platforms and OR mapping tools (these are emerging) alone are two very relevant topics for discussion in the space.
Floyd also made another good point in saying that it's not that .NET community doesn't have much material to cover, but that its hot points are different. The dearth of open source frameworks is just such an issue - people are interested in looking into this from a .NET perspective and they really need a forum that can act as a staging area for that discourse. TSS.NET will provide that in time.
And absolutely everyone has given up trying to figure out when Longhorn is coming out. ;-)
Mike
|
|
|
Message #148059
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
that's like
baptist saying, "catholics aren't christians" because "it's by faith alone".just about everything gets tangled up in religious zealotry. oh wait, it was suppose to be a joke. too bad that is the most annoying part of both java and .NET communities. By orders of magnitude, this exist more so in the java world. Whenever you see a topic on tss.net and tss.com that favors .Net in any way. The tss.com guys come from everywhere making broad baisless statements about .Net and .Net developers. But on tss.net hardly a whisper even when it's the reverse. Another reason why I would never want to work in a java shop, bond to be a zealot in the group who will get on my nerves.
Like some other responses to this, I can confirm that many .NET developers I know do not wander off outside the realm of Microsoft - including technology forums. For them, there is no solution outside of what Microsoft says/does/offers.
You might see a Java developer over here because many do roam around in various forums regardless of the vendor or technology to find solutions or just to learn out of custom. Many Java developers develop with .NET as well.
There are, however, .NET/Microsoft zealots out there! Make no mistake about it :) You will see .NET/Microsoft zealot(s) amongst the top 5 posters on the Java side of this same forum.
|
|
|
Message #148079
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Perhaps .Net devs have different hangouts
I don't for one second believe that the majority of .Net devs that would describe themselves as serious get ALL of their architectural guidance from MSDN simply because architectural guidance from MSDN is only a very recent phenomenon. MSDN is simply the best place to go to get technical answers and articles on .Net (and simply cannot be beat in this regard, AT ALL) Besides what I see is, if a MS provided solution doesn't quite cut it, the community looks elsewhere. EIF not too great? Ah well, use Log4Net. Oh, Team System is only for big shops? *shrug*, lets use CC.Net. So perhaps culturally we're just not as vocal about this stuff as Java devs are, but we;re not asleep at the wheel, believe that.
|
|
|
Message #148085
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Perhaps .Net devs have different hangouts
I don't for one second believe that the majority of .Net devs that would describe themselves as serious get ALL of their architectural guidance from MSDN simply because architectural guidance from MSDN is only a very recent phenomenon. MSDN is simply the best place to go to get technical answers and articles on .Net (and simply cannot be beat in this regard, AT ALL) Besides what I see is, if a MS provided solution doesn't quite cut it, the community looks elsewhere. EIF not too great? Ah well, use Log4Net. Oh, Team System is only for big shops? *shrug*, lets use CC.Net. So perhaps culturally we're just not as vocal about this stuff as Java devs are, but we;re not asleep at the wheel, believe that.
I hope I didn't give the wrong impression that .NET developers are asleep. I was attempting to point out cultural differences. This mainly is the result of company culture and how the CTO/CIO thinks.
If the CTO says, "make sure you look at what microsoft recommends," or "don't give me non-microsoft solutions" what can a developer do? I've met a few CTO's who were that way. Of course they are also clueless monkeys, who got the job because they went to college with the CEO who is also clueless. Many people in those cases simply don't bother looking outside MSDN, or Technet. Why bother really? If I know the CTO isn't even going to consider it, I'm better off figuring out how to solve the problem with what I have instead of proposing an open source solution. The only thing I'll get from a clueless CTO is grief and a pink slip.
since I mainly do contract/consulting work, I always speak my mind. It's gotten me in trouble in the past, but I feel it is better to provide a clear explanation of the potential solutions than play stupid. In some cases, the CTO's took the feedback seriously and corrected the problems. But most of the time, no amount of careful analysis and research can fight bad management.
In my experience, the java shops I've worked at were already running a mix of platforms and languages, so their mindset was already "choose the right tool." Other people will have different experience and my experience has taught me the thing to look for in a company is solid technical managers. without that, it's just a huge blackhole regardless of the language or platform they use.
|
|
|
Message #148086
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Perhaps .Net devs have different hangouts
This is a very good point. The typical mistake made when comparing the two communities is to judge each against the characteristics of the other. This quickly degenerates into "they don't do what we do, so therefore they do nothing" (admittedly, this is almost always a J2EE claim in my experience).
Granted, many are more open-minded than that, but a lot of people forget that there's a valid reason why some fundamental differences exist between the value systems of each community. J2EE and .NET camps have different sets of considerations to prioritize for development and delivery. For example:
- J2EE might have to support a portability requirement (and if not explicitly required this might be a standard operating procedure). This typically leads to a lot of attention on plumbing, app server choice, and other infrastructure that some perceive as the computer science, engineering-oriented approach. It's a necessity for the world they live in.
- .NET will (almost) exclusively support the Windows platform. With such a tightly coupled relationship, a lot of devs/architects eschew worrying too much about the plumbing, because they often don't have to, in favor of getting everything they can out of the platform they're married to. This leads to different sweet spots for the .NET community.
There are a couple of disruptive factors pushing the .NET community now more toward where the J2EE community has had to be for years:
-The advent of multiple platform support and opportunity through such open source programs as Mono and (to a far lesser degree) Portable.NET, and a growing interest in open source software projects.
- The efforts of Microsoft itself to bring the architecture level out as a discipline that some of its developers can adopt and follow. Granted, some of the best .NET architecture guidance can come from sources other than MS, but it brings many to the table for introduction to the architecture world.
|
|
|
Message #148097
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Perhaps .Net devs have different hangouts
This is a very good point. The typical mistake made when comparing the two communities is to judge each against the characteristics of the other. This quickly degenerates into "they don't do what we do, so therefore they do nothing" (admittedly, this is almost always a J2EE claim in my experience).
Since I don't have an CS or EE degree and I've never talken a college programming course, I can say I've seen both sides. When I first started out, I mainly used IIS/HTML/ASP/CGI/Apache and focused on client side applications. Back then, I thought along those lines about people who did server stuff out of ignorance. In my mind, servers were no big deal because I was using IIS. I didn't have to think about the internals of how a server manages applications, compiles pages and pools objects. Once I got into server side development, I realized just how complex high performance/high availability server apps are. The reality is that the problems are so different that most of the time the two camps totally misunderstands the other. I can do things in a client side app that is inefficient and it won't really matter. If I do the same thing on a server application it matters a lot. to use an analogy, it's like the two camps speak different languages, but they think they're speaking the tongue. Kinda like in britain, chips are fries and crisps are chips.
Granted, many are more open-minded than that, but a lot of people forget that there's a valid reason why some fundamental differences exist between the value systems of each community. J2EE and .NET camps have different sets of considerations to prioritize for development and delivery. For example:- J2EE might have to support a portability requirement (and if not explicitly required this might be a standard operating procedure). This typically leads to a lot of attention on plumbing, app server choice, and other infrastructure that some perceive as the computer science, engineering-oriented approach. It's a necessity for the world they live in.- .NET will (almost) exclusively support the Windows platform. With such a tightly coupled relationship, a lot of devs/architects eschew worrying too much about the plumbing, because they often don't have to, in favor of getting everything they can out of the platform they're married to. This leads to different sweet spots for the .NET community.There are a couple of disruptive factors pushing the .NET community now more toward where the J2EE community has had to be for years:-The advent of multiple platform support and opportunity through such open source programs as Mono and (to a far lesser degree) Portable.NET, and a growing interest in open source software projects.- The efforts of Microsoft itself to bring the architecture level out as a discipline that some of its developers can adopt and follow. Granted, some of the best .NET architecture guidance can come from sources other than MS, but it brings many to the table for introduction to the architecture world.
It does appear MS is serious about interoperability and is working with SUN on that. My bias take on it is MS finally accepts that windows can't replace the old mainframes and probably never will. After all of us are dead, mainframes will still be around, but windows in its current form probably will not. Since windows is so much cheaper, replacing it is easier than replacing a big expensive mainframe. Kinda like re-landscaping your yard versus re-building the Capital building in D.C.
|
|
|
Message #148238
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
wow, that makes a lot of sense
I know only 2 of them look outside MS resources for information. The rest happily eat what MS sells. I would say the smae is true on the java side, how many java developers you know swallow everything IBM WebSphere BEA or Oracle has to offer without considering other lightwieght frameworks or app servers that could do the job. First of all, as you mentioned, it’s IBM, BEA, SUN, Oracle (4 not 1) I’d add couple of open-source alternatives. Second – from my experience developers tend to go with $X000- $XX000/CPU app. servers for lightweight apps only if it’s corporate standard and licenses was already paid anyway.
I probably live in other world because developers around me came to java from wintel or unix world (or both), hence beside Java for right task will consider .NET/Perl/you name it. Probably there a devs. Who wrote first “hello world” in VB.NET or in BEA/IBM wizard and strictly stick to these tools, but in current job market I don’t see much vacancies for them.
|
|
|
Message #148245
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
like I said, luck
I know only 2 of them look outside MS resources for information. The rest happily eat what MS sells. I would say the smae is true on the java side, how many java developers you know swallow everything IBM WebSphere BEA or Oracle has to offer without considering other lightwieght frameworks or app servers that could do the job.
First of all, as you mentioned, it’s IBM, BEA, SUN, Oracle (4 not 1) I’d add couple of open-source alternatives. Second – from my experience developers tend to go with $X000- $XX000/CPU app. servers for lightweight apps only if it’s corporate standard and licenses was already paid anyway. I probably live in other world because developers around me came to java from wintel or unix world (or both), hence beside Java for right task will consider .NET/Perl/you name it. Probably there a devs. Who wrote first “hello world” in VB.NET or in BEA/IBM wizard and strictly stick to these tools, but in current job market I don’t see much vacancies for them.
I consider it luck. The java shops I've worked at the developers look around for solutions. What does seem to be consistent is shops that aren't used to mixed environments don't look outside one vendor. Companies that run mixed environments are in the habit looking for different options and picking one. Given that most of my experience has been in mixed environments, luck would have it that I haven't had to work with really bad java developers.
I've heard plenty of horror stories from friends who work in single environment shops. But that's second hand and not first hand experience. Even then, I would say single versus mixed environment is a good indicator of whether a company has a policy of looking else where for solutions. I still think it is a company culture issue. What ever the top managers prefer ultimately ends up defining the culture of a company.
|
|
|
Message #148437
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
too much ignorance out there
The reason why we (the .NET developers) do not look outside Microsoft for advice is:
1) first, because there is no consistent advice out there, only a lot of contracting opinions.
2) second, even if they sometimes agree upon something (like "EJB is the best thing since sliced bread"), it usually turn out a terrible disaster.
3) third, the Java developers have absolutely no idea what can be achieved with Microsoft tools ("MS SQL Server is a replacement for Access":)
Microsoft is superior
Regards
Rolf Tollerud
|
|
|
Message #148447
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
too much ignorance out there
Of course you would know how to build a compliance system that handles 200K transactions a day that have to go through rigorous financial compliance testing :)
just joking. I do know of systems build with C++ on windows that can handle that kind of load in batches, just not real-time. whereas I do know of atleast 4 production java systems. One of them can handle up to 5% of the trading volume of NYSE.
|
|
|
Message #148505
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
MS has nothing to learn in the Java camp
It is not so long ago when 95% of all new systems was done with Microsoft tools and technology. After a all time low under the "Big Java EJB Servers" high-days, for the moment the figure is 54%. For anyone that can extrapolate a graph and know about Avalon/XAML there is no problem to foresee the future.
No matter how you shout, cry, beg, plead, threaten, that is going to happen.
And about my first comment "only a lot of contracting opinions" I should have been more precise; "chaos" or "panic" is the right term to use.
Microsoft is so far ahead that using Java and Java tools feels like going 10 years back in time.
Regards
Rolf Tollerud
|
|
|
Message #148516
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
no lack of hyperbole
I can always count on you for extravagant hyperbole with absolutely no hard facts to back anything up. Not everyone things the way you do and most people are actually pragmatic. that means only a small vocal minority resort to shouting and ranting and calling names.
I enjoy playing devil's advocate, but I atleast try to back it up with cold hard facts. You might want to try that one of these days.
|
|
|
Message #148524
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
hyperbole?
I can always count on you for extravagant hyperbole with absolutely no hard facts
Allow me to laugh. No amount of facts bites on the hardened Java zealot; after all he has learned to live with reality by denial and is master of squirmy excuses. In spite of more that a dozen benchmarks trouncing J2EE for example, all they say is: "all of them is manipulated", "bought by Microsoft".
No, the reason I put this posts is not to try to convince anyone, it is for my own sake. My favorite sentences are namely "What did I say" or "I told you so".
Go back 2-3 years in TSS.com and discover that I was the first to doubt the viability of the Big Java EJB Servers.
So I can say "What did I say". And I look forward to do it once again in the future. :)
Regards
Rolf Tollerud
|
|
|
Message #148547
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
hyperbole?
I can always count on you for extravagant hyperbole with absolutely no hard factsAllow me to laugh. No amount of facts bites on the hardened Java zealot; after all he has learned to live with reality by denial and is master of squirmy excuses. In spite of more that a dozen benchmarks trouncing J2EE for example, all they say is: "all of them is manipulated", "bought by Microsoft".No, the reason I put this posts is not to try to convince anyone, it is for my own sake. My favorite sentences are namely "What did I say" or "I told you so".Go back 2-3 years in TSS.com and discover that I was the first to doubt the viability of the Big Java EJB Servers. So I can say "What did I say". And I look forward to do it once again in the future. :)RegardsRolf Tollerud
so how long are you going to use that excuse? Like you said, those that are blind to the flaws of a given platform and system aren't going to change their minds. But for everyone else, given sufficient proof, they can be reasoned with.
When I set about figuring out how the london stock exchange achieved their performance I wasn't out to prove .NET sucks. On the contrary, I was given the job of figuring out how to reproduce it since management wanted to maximize performance. On a personal level, I did it because I wanted to figure out exactly what they did and reproduce it reliably. After several months of intense research and testing, I was able to reproduce similar numbers. As I said in my original analysis, what they achieved was very impressive, but it's necessary to qualify and quantify exactly what the system is doing. Making BS marketing statements like .NET can handle 2K transaction per second is 1/4 truth. The truth is .NET can handle large volumes of data and provide some significant benefits, if you know what you're doing.
Obviously no marketing department is going to point out their statements are misleading. Not oracle, sun, ibm or microsoft. Therefore, it is up to developers to filter the BS from what is real and challenge the manufacturers. My guess is eventually .NET will provide a stateful application container in the next 5-15 years. Some things are bloated and slow. There's no getting around that. For everything else KISS is really a good model.
Personally, I'm not convince every possible application out there can be satisfied with a simple stateless model that is light and fast. I've seen enough trading and financial applications to know that there are many cases where you really do need a stateful approach. It might be that EJB is the worse possible realization of a stateful container, but I've never written an EJB server and won't bother making claims I can't back up. I do understand how to get around and avoid using stateful approaches, but at some point, the problem of reconciliation has to occur. Be it in a stored procedure with triggers, COM+, EJB, MOM or the flavor of the month.
Then again, you consider my arguments BS, so not like this will have any effect :) By the way, I remember plenty of people screaming EJB was bloated and still born way before you ever said a thing on TSS. As far back as 99/00 I knew plenty of people claiming EJB was too hard and tooooo bloated.
I could resist addressing your other points. If non-microsoft world is so lost, why is it that COM+ borrowed from Tuxedo. Indigo is a nice advancemnt, but it is borrowing from technologies pioneered on mainframes. In a recent TSS.NET thread, ASMX will support interfaces, which has been the preferred model in J2EE.
My point is to borrow and learn from all sides, be it your favorite or not. There's always something to be learned, even if it doesn't apply to your application of current neds.
|
|
|
Message #148558
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
for the record
There are plenty of ideas J2EE borrowed from .NET, so it's not like Sun, IBM, Microsoft and Oracle really believe their own marketing hype either. It's all part of the game to win over developers.
my soul is not for sale to any of them, maybe a timeshare or 12 month lease. then again, if they offered me 50 million for my soul, I'd probably go for it, since I don't believe in hell/heaven anyways.
|
|
|
Message #148559
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
art is long, life is short
Please Peter I never said that your arguments is BS. On the contrary I would say that you are unusually well informed for a Java developer. And I do appreciate that you remember my habits :) (always using stateless servers)
But you argue that that MS technology is not fit for large (many trans/per second). OK. Can you possible give me an estimation of what is the limit in your opinion? For a cluster of 5, 10, 50 and 500 Intel processors? You know of course that Google use over 100000 computers.
Anyhow I can not understand this preoccupation with only large large, big big applications. I am sure that money wise, the really big applications stand for no more 0.01 % of the IT-budgets in the world.
When we go back in time we can see in retrospect that MS was able to obtain 95% of all new development thanks to a total paradigm shift. The customers (users) just prefered the superior user-interface. When MS lost their hegemony is was again because of another paradigm shift. And now it is time for yet another, that is written clearly in the sky, can’t you see it?
I, we, the developers want everything, (to eat the cake and still have it). To develop in a "real" language not Javascript, to have access to all native API and still have the power of XML/HTML/CSS. Silent install without any hassle and click-once installations.
And Microsoft listen and will give it to us with Avalon/XAML. It is impossible to research everything. I limit myself to MS because there is the bleeding edge.
Regards
Rolf Tollerud
|
|
|
Message #148562
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
art is long, life is short
Please Peter I never said that your arguments is BS. On the contrary I would say that you are unusually well informed for a Java developer. And I do appreciate that you remember my habits :) (always using stateless servers)But you argue that that MS technology is not fit for large (many trans/per second). OK. Can you possible give me an estimation of what is the limit in your opinion? For a cluster of 5, 10, 50 and 500 Intel processors? You know of course that Google use over 100000 computers.Anyhow I can not understand this preoccupation with only large large, big big applications. I am sure that money wise, the really big applications stand for no more 0.01 % of the IT-budgets in the world.When we go back in time we can see in retrospect that MS was able to obtain 95% of all new development thanks to a total paradigm shift. The customers (users) just prefered the superior user-interface. When MS lost their hegemony is was again because of another paradigm shift. And now it is time for yet another, that is written clearly in the sky, can’t you see it?I, we, the developers want everything, (to eat the cake and still have it). To develop in a "real" language not Javascript, to have access to all native API and still have the power of XML/HTML/CSS. Silent install without any hassle and click-once installations.And Microsoft listen and will give it to us with Avalon/XAML. It is impossible to research everything. I limit myself to MS because there is the bleeding edge. RegardsRolf Tollerud
I was joking by the way. My perspective is totally bias obviously, since I prefer to work big systems that have to support thousands of concurrent transactions/queries. At this point, I would say my knowledge of the small scale application is lacking. The kinds of systems I am talking about have to support any where from 2-5K concurrent queries as in "they really are running the queries in parallel." Loads like these it honestly doesn't matter if you're using Oracle, Sybase, DB2 or Sql Server since it's a hardware problem.
horizontal scaling doesn't work well in these cases, but honestly most small/medium business never see these kids of insane load. The easiest way to solve these problems is with vertical scaling and really big expensive systems. Though I'm sure Oracle will tell me I'm smoking crack and claim Oracle 10g can handle these kids of loads in a massive blade cluster and horizontal partitioning.
I've never used 10g in that kind of setup, so I'm skeptical until I see it. I do know someone who is moving aggressively to 10g in a cluster setup, since they have insane loads 24/7. I can't go into detail, but their data center is the size of 3 football fields. When I heard the kinds of loads they are handling, I dropped my jaw. I know a few more people with some bleeding edge experience with systems even larger. It's mostly luck that I've been able to meet these individuals and have in-depth consersations about actual real world applications and how we solve different problems. I don't doubt Oracle can get a massive cluster of 500 blades to handle 5K concurrent queries, but I haven't see it. I'm sure someone out there is trying it now and laughing at my remarks :)
Google has done a phenominal job of getting a cluster of 100K servers to work seamlessly to the end user, but they did have to build GFS to do it. Honestly the big market is brutal and I wouldn't want to compete in that space. Microsoft so far has ruled the SMB space and will do so for the next few decades unless someone manages to invent a silver bullet.
On the client side, it's rather daunting to beat microsoft. I'm not sure anyone is up to the task, but since my focus is on server side, most of the bleeding edge grid/distributed computing stuff has been in linux/unix world. Microsoft research has been working very hard, but it's going to take several years for those advances to reach the end user.
Javascript is evil, but it's a fairly complete scripting language. Just not something one would want to use to write a photoshop like application. I mention that because I've had people ask me that in the past. Talk about mis-use of technology.
|
|
|
Message #148629
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
to be dragged, kicking screaming, into the future
Google has done a phenominal job of getting a cluster of 100K servers to work seamlessly to the end user, but they did have to build GFS to do it
Well but it works, so what.
You can not deny that the trend in large systems is Intel clusters. And BTW it was not so long ago that Intel architecture was bashed in TSS, "Not fail safe unfit for large systems". In fact it is possible to compile a long list of former bashed MS/Intel technology that now is market-leaders. Remember Pocket-PC! It is as the public opinion has no memory.. :)
Anyway I think you have misunderstood me, I am not against large systems, I am against (and please read carefully) Crappy user interfaces. Look for instance on Googlies little pearl, Gmail. A nice little application in spite of being done in Javascript. Blowing all other Web-based email clients out of the water. That shows that "Large Systems" and "Quality User interface" are not necessarily orthogonal.
Most interface to "Big Java EJB Servers" today is a joke, more often than not is just a list from one table on the screen where you can edit, delete or change one post at a time. It is as they have worked so much on the serverside so there are no power left at the clientside. It is my firm opinion that in the coming year you have to spend millions at the clientside in Enterprise business. And that is good news for us developers that see ourselves as creators, artists if you like. How many man-hours do you think has gone into Eclipse by this time do you think?
Avalon/XAML and Indigo will change everything at the MS side, and the rest of the world will follow eventually. As Usual.
Regards
Rolf Tollerud
|
|
|
Message #148641
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
to be dragged, kicking screaming, into the future
Google has done a phenominal job of getting a cluster of 100K servers to work seamlessly to the end user, but they did have to build GFS to do itWell but it works, so what.
There was a recent interview with one of the top google guys. The "so what" is that multiple drives fail every single day. To get around that, there are multiple copies. As google put it, without writing GFS, they wouldn't be able to scale or maintain the same level of reliability. Yahoo also has an insane number of servers. The trend is fairly clear, as grid matures, more people will use it for new projects. Whether that means replacing existing system is another story.
You can not deny that the trend in large systems is Intel clusters. And BTW it was not so long ago that Intel architecture was bashed in TSS, "Not fail safe unfit for large systems". In fact it is possible to compile a long list of former bashed MS/Intel technology that now is market-leaders. Remember Pocket-PC! It is as the public opinion has no memory.. :)Anyway I think you have misunderstood me, I am not against large systems, I am against (and please read carefully) Crappy user interfaces.
I've seen my share of bad UI, but that has nothing to do with stateful application servers or even servers at all. That's just plain bad design and bad specs.
Look for instance on Googlies little pearl, Gmail. A nice little application in spite of being done in Javascript. Blowing all other Web-based email clients out of the water. That shows that "Large Systems" and "Quality User interface" are not necessarily orthogonal.Most interface to "Big Java EJB Servers" today is a joke, more often than not is just a list from one table on the screen where you can edit, delete or change one post at a time. It is as they have worked so much on the serverside so there are no power left at the clientside. It is my firm opinion that in the coming year you have to spend millions at the clientside in Enterprise business. And that is good news for us developers that see ourselves as creators, artists if you like. How many man-hours do you think has gone into Eclipse by this time do you think?Avalon/XAML and Indigo will change everything at the MS side, and the rest of the world will follow eventually. As Usual.RegardsRolf Tollerud
The application I know of aren't simple one table views. If anything, in the financial world, they're real-time blotters and monitors written in Swing. Things like historical stock trends, graphs of a securities performance, performance of a portfolio and the current market trend. Doing these kinds of things in HTML is horrible and I totally agree. It's just completely backwards and a waste of time. But to be fair, I have seen some very impressive java applets for stock analysis, which blow me away. It allowed you to zoom into a specific time period for a stock and do all sorts of cool analysis. It was also wicked fast. I have no idea how they managed to do it in an applet, but I was impressed. If I had reproduce the same functionality, I'd go with swing simply because I don't have that level of applet expertise.
I would concur there is a trend back to rich clients. To my limited knowledge, many shops in Boston started the move 2-3 years ago, so it's not new. The difference is that everyone pretty much wrote their own framework, so it will be nice to have a consistent framework like XAML. that will avoid everyone re-inventing the wheel, since the financial firms are very secretive and don't go tell the others what they are doing. What I see, which may be totally wrong is J2EE will get more entrenched on the server side and .NET will get more entrenched on the client side.
This is why I feel it's important to understand and know both. when I talk to my colleagues in the financial sector, a common statement is "we're building a new system that has to last atleast 20 years and possibly 30 years, so it has to be extensible." By extensible they usually mean reduce the cost of maintenance from the typical 80% down to 30%. That is one factor driving UI design also. Often making the UI pretty isn't high priority honestly. More often than not, people will say, "let's get the functionality rock solid and worry about pretty later." I don't think that mentality will change honestly. For better or worse, that's what it is.
|
|
|
Message #148660
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
"pretty"?
Do not confuse "pretty" with functional. In fact web apps are often nice aesthetically, but dysfunctional. With good UI-interfaces I mean tools that allow a user to get four times as much work done in the same amount of time compared to a typical "pretty" web application.
I have seen some very impressive java applets for stock analysis
Of course it is possible to do good (in the meaning functional) IU with Java applets or standalone-install-on-demand Swing apps. But Sun blew that when they sued
Microsoft. Now it is C# applets and .NET that counts. And even more so when Avalon/XAML kicks in. Java is a legacy platform. (They can join the hordes of unlucky slaves that are still maintaining Cobol)
It is called, "To saw of the branch you are sitting on".
Regards
Rolf Tollerud
|
|
|
Message #148666
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
"pretty"?
Do not confuse "pretty" with functional. In fact web apps are often nice aesthetically, but dysfunctional. With good UI-interfaces I mean tools that allow a user to get four times as much work done in the same amount of time compared to a typical "pretty" web application.I have seen some very impressive java applets for stock analysisOf course it is possible to do good (in the meaning functional) IU with Java applets or standalone-install-on-demand Swing apps. But Sun blew that when they sued Microsoft. Now it is C# applets and .NET that counts. And even more so when Avalon/XAML kicks in. Java is a legacy platform. (They can join the hordes of unlucky slaves that are still maintaining Cobol)It is called, "To saw of the branch you are sitting on".RegardsRolf Tollerud
that's a good point. but from experience doing web design in the early 90's, everyone and their pet dog has a different definition of functional and pretty. then again, that's why I moved to server side. my own bias obviously.
you're comments are always so color.
|
|
|
Message #148673
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
a powerful concept
It is all about money. When I can deliver an app that allows the users to do their work in a better way I get the contract. That is my method (the UI) that outcompetes the competition. Of course if my application in addition is faster and scale better that is no disadvantage. That I perhaps not can accept the 0.01 biggest projects in the world I can live without, because neither can the J2EE guys.
In fact, "Very good UI, Stateless Servers, Intel Clusters, SOA" is a very powerful concept and I am sure you have not heard the last of us yet, even in the financial sector.
Regards
Rolf Tollerud
|
|
|
Message #148678
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
As LaoTzsu would say
the only constant is change. in twentyfive years 80% of the people on TSS.com and TSS.net maybe using the next big thing and most likely it will be something different.
but here's my goal. If I really understand the underlying principles and low down details, it won't matter what language is the flavor of the day. It also won't matter which UI toolkit is hot, if I know the fundamentals of how widgets render graphics, approrpriate use of events and methods for lazy loading/lazy evaluation.
when it comes time to build something, it's just a matter of learning a different syntax and the quirks of that new language. I don't doubt change is on the way, but if I separate the hype from the physics of how things really work, I make it easier for myself to adapt. Being a consultant, adaptability is more important than preaching to my client. I refrain from preaching and prefer to listen to my clients first before opening my loud mouth :)
|
|
|
|
Hot threads
Hot threads
Hot threads
|
More hot threads
More hot threads
More hot threads
|
|
Speaking at EclipseCon 2006, Java developer and independent consultant Madhu Siddalingaiah compared Microsoft's Visual Studio IDE to the open source development environment of Eclipse.
(33 comments,
last posted
July 13, 2009)
In this tech talk, Microsoft's Peter Provost talks about the design of the Composite UI Application Block and how the p&p team has led Microsoft in the adoption of Agile methodologies.
(0 comments,
last posted
April 17, 2006)
Chapter 4 of Framework Design Guidelines, titled "Type Design Guidelines," presents patterns that describe when and how to design classes, constructs and interfaces. In this chapter, Abrams and Cwalina divide types into four groups and discuss the do's and don'ts of type design.
(2 comments,
last posted
July 07, 2006)
Paul Ferrill caught up with prime open-source .NET applications driver Miguel De Icaza at Novell's BrainShare conference last week. They discussed the status of Windows Forms for Mono (it's coming along) and VB.NET for Mono (it looks like it's out).
(5 comments,
last posted
March 30, 2006)
In this tech talk, Microsoft Visual Studio architect Jack Greenfield discusses the company's approach to Domain-Specific Languages, or DSLs, and the part they play in software factories.
(0 comments,
last posted
March 15, 2006)
More hot threads »
|
|
