__/ [ BearItAll ] on Tuesday 23 January 2007 11:20 \__
> Roy Schestowitz wrote:
>
>> __/ [ BearItAll ] on Tuesday 23 January 2007 09:36 \__
>>
>>> Roy Schestowitz wrote:
>>>
>>>> Can Red Hat rival Microsoft?
>>>>
>>>> ,----[ Quote ]
>>>> | In the previous post we saw that Oracle's Linux initiative has so far
>>>> | failed to make a serious dent in Red Hat's business or even in its
>>>> | stock price. Red Hat is actually worth slightly more today than it was
>>>> | when Larry Ellison launched his apparently not-so-scary RHEL clone the
>>>> | week before Halloween.
>>>> `----
>>>>
>>>
>>> The problem there is that it would be very unlikely to be a quick win.
>>> They are targeting IT who don't jump from contract to contract on a whim,
>>> they can't do their job like that.
>>>
>>> You wait until the current one is running out, and start planning the
>>> next, so many that Oracle are targeting will currently be in contract. It
>>> is when those contracts are running out and you are budgeting for the
>>> next where a large difference in costs will almost certainly make a
>>> difference.
>>>
>>> Particularly if you take into account that you don't actually get much
>>> for your money, you do get tried and tested updates which of cause can
>>> easily be argued is worth the money, because it is so solid that you
>>> could even update the live server (if the pessimist in you would allow
>>> it).
>>>
>>> I paid support for RH, I never once got in touch with them, true that it
>>> was a nice cushion to have but actually when RH did the dirty and
>>> cancelled the previous system part way through the contract. I never got
>>> another one, when I did a new RH server I paid that first year, but never
>>> extended them since that time. It doesn't make the slightest difference
>>> to me. In fact, rather than stay a few versions behind on everything, I
>>> went ahead of what would have been under the support banner. Newer MySQL
>>> and other bits-n-bobs.
>>
>>
>> The fact that you didn't need support is indication of satisfaction
>> though, no? Even when covered by RHAT (now RHT), any problem you may face
>> costs you time (and at worse scenarios --- downtimes).
>>
>
> But I also never had downtime, caused by the OS, with Suse, Mandrake,
> Slack, Debian .....
>
>>
>>> Many IT'ers were very angry with RH when they did the nasty with those
>>> license contracts, when IT'ers get together it still comes up in
>>> conversation.
>>>
>>> Then in the budget meetings when you put down the proposals for the next
>>> year, the numbers stand out a bit, "So Rob what your saying is we can
>>> either pay RH $xxxx or pay Oracle $xxx, but we get the same updates etc
>>> which ever one we go with?", they'll look at you as if your stupid (and
>>> may well be right). Besides, if you pay $xxx then you have $x000 to spent
>>> on other IT toys, I wonder if I can get a Wii through the budget, what do
>>> you think?
>>
>>
>> You need to explain to them that IT workers must get a share of exercise.
>> Yesterday on the Register (maybe Inquirer) there was an article about a
>> guy who lost a lot of weight by playing 'Wii sports'. Rather than
>> companies paying for gym memberships for their staff they could throw over
>> a hundred quid on a box that plugs into a TV. You could take turns and
>> share the machine. :-)
>>
>
> I have copied and pasted that into my budget plan :)
You might want to sed it.
s/wii/enterprise-exercise-machine/
>>> So don't go thinking that Oracle have lost on this one, RH still need to
>>> do something radical to ensure that when renewal date comes that they are
>>> in the list of options.
>>
>>
>> Oracle's product is said to be unreliable and unattractive. maybe this
>> will change in the future. They did a lot of PR (paid for) yesterday. It
>> flooded the RSS feeds that I follow.
>>
>
> They have no choice but to put up a fight, Oracle dbase is a great product,
> many would be very reluctant to move away from it and many would still make
> it the first on the list of possibles for new database contracts,
> particularly for large scale databases, either large in data or large in
> concurant users.
>
> But that is changing, where once even relatively small databases would use
> Oracle, simply because the developer was used to it and trusted it, many of
> those smaller systems are being moved or developed on cheaper alternatives,
> postgresql and MySQL, some are finding that sqlite actually fits the bill
> on many a small database that at one time might have been money in the bank
> for the likes of Oracle. Unidata is feeling a similar pinch for the same
> reasons.
>
> Then when you concider the ease at which postgresql and mysql marry with a
> selection of scripting and programming languages, this makes them very
> powerfull indeed, also very flexible, maybe you would choose perl as a main
> backend for controlling your database and adding functionality, or maybe
> c++, python, Ruby, GLBCC (Basic), it all gives you more flexibility when
> selecting available developers from the job market.
>
> Probably the only thing holding these two back now are their limits in
> connections and shared space. These are good enough for many applications,
> but probably not good enough for large databases with potentially many
> concurent users, banks or large communication systems, unless there were
> built in a tree fashion to distribute the work but that adds to the overall
> cost anyway.
>
> So, on that sort of scale Oracle is still on top, but it is quite clear
> that it will not continue that way as use of postgres and mysql grow and
> the limits are lifted. Someone said in a forum not too long ago that it may
> well be that the only thing holding MySQL off performing well in the the
> database scenarios is likely to be a bit of clever optimizing at compile
> time. I don't know if that is entirely true, but they may well be a bit of
> truth in it.
The thing about OSS is, the more people /use/ it, the more people get
involved in development (sometimes contributions that flows back to the
vendor). I heard this from MuleSource, for example. I see a lot of this in
WOrdPress (been close with the devs since 2004). As it's not very vendor
dependent (GPL licnese), the DB's will only improve. They will improve at a
non-linear pace owing to growth (userbase, 'network effect') and close the
gap. This can of course be generalised. Just look at Beryl/XGL and Linux
virtualistion. Looking at Linux in 2000 (that's when I got introduced to
it... I was only 18 at the time), who would have thunk (sic) and foreseen
this vision? Even prototypes of Apple's and Microsoft's next hits are
uninspiring, IMHO. Just look at KDE 4 usability and conceptual
integration...
--
~~ Best wishes
Roy S. Schestowitz | Kernel panic - more exciting than being /.'ted
http://Schestowitz.com | Open Prospects ¦ PGP-Key: 0x74572E8E
Tasks: 120 total, 2 running, 117 sleeping, 0 stopped, 1 zombie
http://iuron.com - knowledge engine, not a search engine
|
|
