[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: The business case for using a linux client versus a windows client
- To: lchalupa@seelink.org, Atlanta Java Users Group <ajug-members@ajug.org>
- Subject: Re: The business case for using a linux client versus a windows client
- From: Jason Chambers <tooger@bellsouth.net>
- Date: Sat, 21 Dec 2002 11:57:51 -0500
- In-Reply-To: <oprhl6ho1ioim9oc@localhost>
- References: <oprhl6ho1ioim9oc@localhost>
Regarding Java and .NET (actually J2EE versus .NET is probably
a better discussion), it can be summarized as follows:
1. Language and platform
o J2EE is multi-platform (Solaris, Linux, Windows etc.), single language
(Java)
o .NET is single-platform (Windows) and multi-language (C# etc.)
2. Vendor
o J2EE - you have a staggering choice of tools, application servers etc. J2EE
is a marketplace.
o .NET - you may have a choice with tools (although with Microsoft in the
driving seat, they can use this to their advantage - remember Borland's OWL
versus MFC? - which one do you think was better? Which one became the
standard?)
o You certainly don't have a choice on the platform vendor. (Although it will
be interesting to see how Mono pans out www.go-mono.com).
In my opinion, there will be organizations that use one or the other or most
probably both. The promise of Web Services strives to ensure they can at
least be compatible.
Personally, I like taking the skill sets (Java) I have nurtured for developing
web applications, and applying them to developing wireless apps or desktop
apps. The multi-language aspect of .NET scares me a little from a maintenance
perspective. To manage maintenance costs, it is best to reduce the technology
footprint (the languages, tools, platforms etc.).
It comes down to selecting the best tool for the job. Let's face it, Swing is
the not the best tool for developing rich Java clients that run on Windows.
Perhaps .NET clients with a Java middle-tier may pan out to be a good
combination.
Now, onto Linux. Again, our challenges are finding the best tools for the job
while reducing costs. GNU/Linux does some jobs really well and has shown to
be very cost effective. Web serving (apache), file and print serving (Samba)
spring to mind. I'm not sure about the corporate desktop - just yet! Java on
Linux is great for development, but I'm not sure about production. Java is
inherently multi-threaded - it's in the language. Linux does not show the
same support for multi-threading than Solaris. From my experience, Solaris is
the best platform for Java (HP-UX ain't bad either!).
Personally, I enjoy the freedom of GNU/Linux. The analogy is that for Windows,
the hood is firmly welded shut. With GNU/Linux the hood is open for you to
tinker with the engine - if that's what you enjoy (I do). You can configure
everything from the desktop look and feel right down to the Kernel. A good
example is I can configure the kernel by removing modules that don't apply to
my hardware - e.g. If I don't care about USB support, I can remove the module
and optimize my kernel for my hardware. It's not for everyone.
So, back to your question "What is the business case for using a linux client
versus a windows client". Well, the good news is with Java - you have a
choice! And ain't choice a wonderful thing?
Jason
On Wednesday 18 December 2002 13:47, Lee Chalupa wrote:
> Hello:
>
> Thank you to those that stopped by the lounge after the AJUG meeting last
> night. I value the opportunity to talk
> to my associates. I hope we can make this a regular event.
>
> One topic that came up was the religion of Java versus .Net and that lead
> to some further discussions about the linux client versus the windows
> client.
>
> I think we all like to have fun with Bill Hates but I don't think we should
> confuse that funning with our business sense: Java Developers are business
> people too!
>
> So, I need to answer a business question for myself. What is the business
> case for using a Linux client versus a windows client as it relates to my
> role
> as a Java Developer? How will this decision lower my costs, increase my
> sales, make me more marketable, or reduce my risks? What other activities
> might I spend my time on that would provide greater return on my investment
> of time and money?
>
> As always, if I get significant feedback, I will compile the results and
> publish the findings in a concise format to this mailing list.
>
> Thanks for your time.
>
> Lee Chalupa
> lchalupa@seelink.org
> 770 381 2377