Loading ...
Sorry, an error occurred while loading the content.

Re: [LoadRunner] Re: It continues...

Expand Messages
  • John
    thats sounds great that you have done so much of research, i think mercury has developed siebel-web for siebel crm application, you have mentioned that you
    Message 1 of 38 , Sep 1, 2007
    • 0 Attachment
      thats sounds great that you have done so much of research, i think mercury has developed siebel-web for siebel crm application,
      you have mentioned that you have done that for read only scenarios, can we assume that you are just fetching data,




      chandna <chandna@...> wrote:
      Me think Mercury has fooled a lot of pople for a lot of time
      including folks internally.

      I have been able to work with Siebel(so far) using web alone. I have
      done experiments and see what additional stuff that Siebel-Web does
      and found it to be quite trivial.

      I will send a cheque of $100 to anyone (I mean the first one to it)
      who can give me a substantial reason for going for Siebel web.
      Something that can convince a company to shell out $50K.

      Now mind it that when I say of Siebel Web I distinguish from ability
      to use Siebel provided dll - ssdtcorr.dll which could be useful on
      occasions. However ability to call this dll is not same as
      worthwhileness of Siebel Web protocol. In fact I am calling this dll
      from Web protocol.

      Let me share the experiments I did to compare the moving parts.
      1. Web alone - hard coded
      2. Web alone- completely manually correlated
      3. Web + Vugen native Siebel autocorrelation
      4. Siebel Web alone
      5. Siebel Web + Vugen native Siebel autocorrelation
      6. Web + Siebel provided correlation dll
      7. Siebel Web + provided correlation dll
      There has been some amount of manual intervention with autocorelated
      scripts.

      The comparison was done for script level with Vugen and then through
      load test using Controller. These have been done for read only
      scenarios so far.

      I also read Siebel architecture documents pertaining to UI (physical
      & logical) interfaces.

      With scenarios that I have handled so far, name of the game is -
      what is called 'star arrays' or 'array of data'.

      It's funny but most hard coded values, such as of row ids are fine
      with Siebel and it does not even impact performance.

      --- In LoadRunner@yahoogroups.com, John <hyklass95@...> wrote:
      >
      > Ashish,
      > Not exactly,stick to the protocol , all i have to say, mercury
      is not a fool to have siebel-web as protocol also its good that you
      prefer siebel web over web for siebel and its great that you do
      manual correlation,
      >
      > Ashish Tyagi <ashish_tyagi123@...> wrote:
      > John,
      > Does this mean that those who can handle manual correlation can
      > manage Siebel scripting without a Siebel-Web license? Actually I
      > have done performance testing on Siebel using Siebel-Web protocol.
      I
      > dont prefer auto-correlation so most of the correlations were done
      > manually.
      > Regards,
      > Ashish.
      >
      > --- In LoadRunner@yahoogroups.com, John <hyklass95@> wrote:
      > >
      > > Secondly you don't need to understand what each web_url is
      doing.
      > As
      > > long as the request is correct we don't need to know what
      exactly
      > > that request is out to achieve. As an example we correlate
      > aplenty,
      > > but do not understand the significance of those parameters. As
      an
      > > example we do not know 'SWEACn, why is SWEC needed, why is SWETS
      > > needed, how is that when you comment out SWETS then also
      > application
      > > works?' etc.
      > >
      > > how will you identifly your request is correct if you cannot
      > read the what web_url stands for?Most of the guys here except few
      > uses the auto correlation for handling correlation and if it fails
      > to take care of things, then your understanding of the statements
      > will help you to identify the target.
      > > i dont remember i have ever used siebel correlation library
      > before and also i am not sure how much HP Guys know but i do know
      > that siebel is vast language in itself.
      > >
      > >
      > >
      > >
      > > chandna <chandna@> wrote:
      > > John,
      > > First question is who decides what's right protocol. Large
      > > population inside HP itself is unable to answer what is that
      > Siebel-
      > > Web does that we can't do using Web. Now this is not about
      Oracle
      > > provided Siebel dll-ssdtcorr.dll.
      > >
      > > Secondly you don't need to understand what each web_url is
      doing.
      > As
      > > long as the request is correct we don't need to know what
      exactly
      > > that request is out to achieve. As an example we correlate
      > aplenty,
      > > but do not understand the significance of those parameters. As
      an
      > > example we do not know 'SWEACn, why is SWEC needed, why is SWETS
      > > needed, how is that when you comment out SWETS then also
      > application
      > > works?' etc.
      > >
      > > Now continuing my finding, I find that Siebel correlation
      library
      > > can be called from Web protocol script. So I still don't
      > understand
      > > why does someone has to go for Siebel Web protocol. From my
      > research
      > > I have seen it to provide three extra variables- SWETS,
      Converted
      > > URL and SWEC. One doesn't have to pay top $ just to get this
      > > facility. Anyone could do it manually as well.
      > >
      > > Now also to demystify the black magic of Siebel correlation
      > library,
      > > what it seems to be doing is that takes as input the entire page
      > and
      > > the creates array of values that we leverage in subsequent
      > requests.
      > > Even Loadrunner provided Siebel correlation does that. Though it
      > > does seem that when array of value gets large Siebel dll is
      better
      > > being more stable and doesn't have any memory leaks.
      > >
      > > But please note this song of praise is for Siebel dll and not
      for
      > > Siebel Web protocol. They can be used independently or together.
      I
      > > am trying to understand how does Siebel Web help anyone.
      > > regards,
      > > Vikram
      > >
      > > --- In LoadRunner@yahoogroups.com, John <hyklass95@> wrote:
      > > >
      > > > if anyone is using not a right protocol, then you need to
      > > understand what those lines stand for ?.
      > > > if you are using http in place of siebel web than you need to
      > > know that each web_url doing or else stick to protocols
      > > >
      > > > you have lot of swe counts coming up without knowing what each
      > > of them are doing .
      > > > sometimes siebels just wont let you know whats its doing and
      > those
      > > help topics in LoadRunner are just good for nothing.
      > > >
      > >
      > >
      > >
      > >
      > >
      > >
      > > ---------------------------------
      > > Luggage? GPS? Comic books?
      > > Check out fitting gifts for grads at Yahoo! Search.
      > >
      > > [Non-text portions of this message have been removed]
      > >
      >
      >
      >
      >
      >
      >
      > ---------------------------------
      > Sick sense of humor? Visit Yahoo! TV's Comedy with an Edge to see
      what's on, when.
      >
      > [Non-text portions of this message have been removed]
      >






      ---------------------------------
      Boardwalk for $500? In 2007? Ha!
      Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games.

      [Non-text portions of this message have been removed]
    • chandna
      Top post, John- very objective. Probably I would have also gone Siebel Web way, if I would not have faced the situation where we had financial crunch. Further,
      Message 38 of 38 , Sep 6, 2007
      • 0 Attachment
        Top post, John- very objective.
        Probably I would have also gone Siebel Web way, if I would not have
        faced the situation where we had financial crunch. Further, I would
        not have espoused Web protocol if there would have been slightest
        difficulty to code/implement. However I have been shouting over the
        top of the roof that, it's just three additional things which Siebel
        Web does. 1. SWEC could be done by correlation also. 2. SWETS code
        can be found on web because many other application also depend on
        time for use as sessionid. 3. URL conversion is also 2-3 line code.

        And good news is that I just implemented an insert scenario in Web
        protocol as well. I can see the records created in app after running
        web script.

        I looked at server logs, database values and placed extensive check-
        points. All that I would have never done if I had Siebel Web because
        then I would have been following the process.The by-product of my
        additional research is also that I have understood how Siebel UI
        works and talks to server and what I can do to make Siebel scripts
        more stable. I discovered bug in ssdtcorr.dll and devise work-around
        for that on basis of my understanding of Siebel. I had to do a lot
        of research because the onus was on me as I proposed doing a new
        way. But it benefitted me in other ways.

        Your post sums up the situation for me - the fear factor.
        'Who knows what will happen if I don't go by the way recommended by
        HP/Mercury? What is the guarantee?' It's scary what all gets sold &
        bought because someone says 'it's recommended and The process'.

        I was hoping the Gurus here wouldn't have any such fears or fear to
        try and challenge what seems irrational. Although it is not very
        unusual to see such fear because if we look at industry, the fixed
        cost(keep the stuff running) versus developmental (new project)
        budget ratio is 70:30. Maximum money still goes in keeping
        the 'lights on' because there is fear factor around what will happen
        to system if I cut on maintenance. It's sad because what some CIOs
        mention is that best guys work on developmental projects and often
        they have to be squeezed out due to cost cutting. With $50K once and
        $10K that we would have to shell out every year on protocol license
        maintenance the business could as well run small projects to do
        security testing, usability testing etc and derive extra ROI on
        software. Folks in IT have to get that business orientation. I
        totally understand that LR itself is overpriced and that we could
        achieve a lot of economy by using freeware like OpenSTA etc. However
        let's not make 'best the enemy of better' as many companies already
        have invested in Loadrunner. If there has to be any caution it is
        around new products that HP would sell which would be in nature of
        add-ons with propositions like productivity (like Ajax protocol) and
        improving software ROI. These need to be carefully evaluated by
        technical community to make sure that we get full bang for bucks. We
        can afford to do that because there is alternative. Maybe for
        something like Remote Desktop protocol we can't because there is no
        alternative.

        Did someone manage to read to this point?..:)
        My previous boss said I write more than what he could read.
        He stopped asking me for detailed reports...:)

        regards,
        Vikram








        --- In LoadRunner@yahoogroups.com, John <hyklass95@...> wrote:
        >
        > vikram,
        >
        > To be honest, i will prefer to stick to protocol to get the
        correct results, if client is not able to afford the price, then i
        can say i have better options with Silkperformer,WebLoad,Jmeter and
        lot of plenty other tools.
        >
        > i have seen that Siebel -web just records the itemdata wrongly
        somtimes skiping the required data, that was way with LR8.0 i quess
        though at that point i did not pay much attention to it reason was
        always i believe doing things correctly as we are hired to get
        results rather than experiment.
        > Maybe you can give a try with update etc scenario and let us
        know.Please note that if you have good coding skills then definitely
        one can go ahead with winsock and do the things,but here if you look
        down as a group of people with varying skills set,your assumption
        does not hold good and utimately productivity goes down.Also i had
        left research long back though was one like you sometimes back.i
        appreciate your work but think anything which works for your
        application might not work with others.
        > Maybe i will contiune this debate once my transition is over.
        >
        >
        > chandna <chandna@...> wrote:
        > My question was not how Mercury sells the protocol as I
        am not
        > discussing marketing strategies. It's from technical stand-point-
        > why should anyone need to buy this protocol if they already have
        > web. The ignorance and confusion sorrounding this just refuses to
        go
        > away.
        >
        > Now read this carefully-
        > "$50K for providing you auto SWETS, SWEC and URL conversion
        handling"
        > That's the value of buying Siebel Web. This is not same as
        > hypothetical case of using Winsock or writing my own protocol. And
        > these three are not more than a day's job
        >
        > And someone please explain me why are Star Arrays are part of this
        > discussion? We are comparing two protocols. Star Arrays are
        handled
        > by LR Siebel autocorrelation as well Siebel Test Automation-
        > ssdtcorr.dll in different ways. However both of them are available
        > with Web protocol as well. Try for yourself and you will find that
        > ssdtcorr.dll can be called and used from Web. So this is
        absolutely
        > out of equation as it is common to both protocols.
        >
        > If you want to know value of Siebel Web do this experiment- Record
        > just using Siebel-Web without using either of ssdtcorr.dll and
        > Siebel auto-correlation. Then record against web with no auto-
        > correlation assistance. Of course Web script will be quite bland.
        > I am waiting to read list of findings on great complexity busting
        > features in Siebel Web script which are sufficient incentive to
        > spend $50K.
        >
        > That articulation is what I had in mind when declaring $100
        > incentive.
        >
        > Now is long answer point to point to John and to Steve.
        > ----------Long answer to Steve-----------------------
        > -My self eulogy is unfortunate but essential requirement to steer
        > clear of profile discussions
        >
        > - I am not tracking h files or F1 parse star array argument
        because
        > there is no need for me to do it. ssdtcorr.dll is doing it for me
        > and so far it seems good.
        >
        > -Winsock etc are specious arguments. I can distuingish water
        melons
        > from grapes. I know effort required in handling count using SWEC
        > correlation, writing SWETS time function and converting URL versus
        > handling it all in Winsock.
        >
        > -How about this more reasonable argument against 'if you are so
        > smart with Loadrunner' - What if I tell you, 'you don't have to be
        > extra smart with Loadrunner to utilize web protocol for Siebel Web
        > application?'
        >
        > -As far as buying from Mercury goes, they are selling it part of
        > Oracle bundle. Of course if you have Siebel Web already there is
        no
        > need to argue, just use the protocol. Question arises for those
        who
        > do not have it.
        >
        > -ssdtcorr.dll free or not is not part of this discussion. I have
        > already acknowledged it's value. But it cannot be equated to
        Siebel
        > Web protocol. It's an add-on this will warp around Web or Siebel-
        Web
        > protocol.
        >
        > -I see personality reference, hypothetical cases and treatise on
        > Mercury marketing techniques as red herrings. A simple recording
        of
        > two scripts to compare and evaluate usefulness could have been
        > better technique to earn $100.
        >
        > ----------Long answer to John-----------------------
        > -John made very valid point that I havent tried insert/update
        > scenarios yet, so it's early. However my response to that is that
        I
        > do not foresee how Siebel Web would do anything different. It will
        > still just do three things extra. Rest still have to be manually
        > handled or through ssdtcorr.dll. Anyway I am about handle an
        > insert/delete script, so I will know.
        >
        > -Maybe my scripts are not complicated enough, but then is Siebel
        Web
        > providing solution to complexity and if so, how? What extra
        facility
        > does it provide at scripting level?
        >
        > -I do not know of Ajax. However we need to distuingish
        productivity
        > from technical feasibility. It is fine having something for
        > productivity if the cost is justified by effort reduction. That is
        > not the case with Siebel Web because everyone mixes up
        ssdtcorr.dll
        > capability with Siebel Web capability. The former is helpful, no
        > doubts.Same goes with Ajax. If a third part vendor gives you a
        plug-
        > in to make your scripting life easy, would you pay for Ajax
        protocol?
        >
        > -I am crying hoarse that Web Siebel is a shallow sheath over Web
        and
        > does nothing special extra. In words of HP it's productivity
        assist
        > and not even technical feasability creator.
        >
        > - At HTTP level, will a Web request be any different from Siebel
        > Web, if the request content is the same? I doubt it, but I can
        > verify.
        >
        > -I am not out here to create a new eco-system proposing few base
        > protocols to handle all protocol. I have reserved focus just on
        > Siebel Web vs Web.
        >
        > -Now the fact Web doesn't record some request that Siebel Web does
        > is indeed news and that's the only worthy argument I have seen so
        > far and this should come close to encashing $100, if you can let
        me
        > know where exactly Web doesn't record. Recording mechanism for
        these
        > two seem very identical. Did you choose right recording options
        when
        > recording with Web- "HTML-A script containing explicit URL only,
        Non
        > HTML generated elements Do not record"
        >
        > I just recorded using Siebel Web and Web. And the recording log
        > header showing Network analysis are exactly identical. The request
        > content also seems similar, though I will spend time on verifying
        > this.
        > -------------------------------Web----------------------
        > Analyzer Module: WPLUS (value=)
        > Analyzer Module: WebBase
        > (value=GetHttpProtocolAnalyzer:api_http_filter.dll)
        > + Network Analyzer: api_http_filter.dll @ GetHttpProtocolAnalyzer
        > Loaded!
        > + Interception Auditors:
        > WinInetWplusInterceptionAudit:api_http_filter.dll
        > Analyzer Module: QTWeb (value=)
        > Analyzer Module: local_server (value=)
        >
        > ------------------------Siebel Web-----------------------
        > Analyzer Module: WPLUS (value=)
        > Analyzer Module: WebBase
        > (value=GetHttpProtocolAnalyzer:api_http_filter.dll
        > + Network Analyzer: api_http_filter.dll @ GetHttpProtocolAnalyzer
        > Loaded!
        > + Interception Auditors:
        > WinInetWplusInterceptionAudit:api_http_filter.dll
        > Analyzer Module: Siebel_Web (value=)
        > Analyzer Module: local_server (value=)
        >
        > regards,
        > Vikram
        >
        > --- In LoadRunner@yahoogroups.com, cheney <socheney@> wrote:
        > >
        > > Your question was why the sweb protocol exists and how
        > > mercury gets by selling it... 100 bean reward....
        > >
        > > So I'll give you a quick argument followed by a long
        > > one. I'll use your own argument: If you are so
        > > awesome with LoadRunner; why aren't you saving
        > > yourself a boatload of money and just using Winsock
        > > for everything? Just a day's work like you
        > > said...right? The reason you are paying for the web
        > > protocol over just, oh, I don't know.... using Winsock
        > > for everything...is basically the same argument I'm
        > > making for why siebel web exists and why it's got that
        > > price tag. I'm sure you could just sift through a
        > > bunch of .dll's and header files and come up with the
        > > web protocol on your own, right? NO?
        > >
        > > So again; you owe me 100 dollars. ha.
        > > I get what you are saying; but your question was how
        > > and why it's marketed...
        > > and you are missing the point... a couple o' things
        > > very key. You clearly don't see yourself as among the
        > > unwashed masses that are incapable of performing URL
        > > encoding; parsing character arrays or noticing that
        > > the siebel count even exists let alone the noticing
        > > how it increments... the very idea that I may count
        > > you as part of the unwashed masses; seems to steer you
        > > off course... so let's assume you know everything
        > > there is to know. (If you did; you'd dig out the h
        > > files and track down that "peculiar" "F1 parse star
        > > array" arguement in the save_param; as well as how the
        > > count is tracked--how and where that stuff is buried
        > > would help you answer your question for yourself; but
        > > you aren't doing that...but never mind that...)
        > >
        > > So, again, the converse; assuming you know everythign
        > > there is to know; and that the correlation dll that
        > > comes with siebel is the whole answer (it's not; and
        > > it's not really free either if you think about it...)
        > > and...AND; keyword being "AND" if you take into
        > > account you were asking what Mercury is doing and why
        > > are they selling it as a separate protocol.. (they
        > > really aren't-- it can be bundled as web.. if you were
        > > planning correctly you could get it all together with
        > > web/winsock and not pay for s-web ala cart...) ...
        > > but .... your question; Why are they selling it
        > > and why is there a market? and I'm handing you the
        > > unwashed masses. The same group of people getting
        > > targeted for "point and click" mode.
        > > Supply/Demand... Cost/Benefit...
        > >
        > > That was the question; this is your 2nd and 3rd
        > > logical answer. Your up to 300 dollars now.
        > > :-)
        > >
        > >
        > > --- chandna <chandna@> wrote:
        > >
        > > > Good now we can start talking about Siebel Web once
        > > > again.
        > > > I am going after Siebel Web because thats what I am
        > > > working on so I
        > > > am hot on it. It's low hanging fruit with huge
        > > > returns ($50K+). I
        > > > don't understand how this illogical proposition
        > > > stayed around with
        > > > so many Gurus such as you minutely inspecting every
        > > > byte of
        > > > Loadrunner products. On this I am driven by
        > > > anatagonism to
        > > > irrational, superstition and feudalism around what
        > > > is right and what
        > > > is wrong.
        > > >
        > > > I am not advocating hard coding and readers of this
        > > > forum have to
        > > > take things with a pinch of salt and read all of my
        > > > post. I report
        > > > facts I have observed and I will reiterate those all
        > > > over
        > > > again.
        > > > a. I acknowledge benefit of LR provided Siebel
        > > > autocorrelation
        > > > b. I acknowledge benefit of Siebel provided Siebel
        > > > Test Automation
        > > > or the ssdtcorr.dll which is an alternative to above
        > > >
        > > > c. I note that both of above work with Web protocol
        > > >
        > > > So this leaves me wondering what is that Siebel Web
        > > > is doing.
        > > > And here are three additional things off the cuff I
        > > > have seen it do-
        > > > SWETS, SWEC and URL Conversion. I would have deduced
        > > > all of this
        > > > myself as well in a days' time and implement in Web.
        > > > So then is this
        > > > what clients are asked to pay $50K for?
        > > >
        > > > $100 offer is still open.
        > > >
        > > > Now for Bangalore and other topic, maybe some other
        > > > time. I do know
        > > > there are some good folks around here as well.
        > > > Hopefully someone
        > > > will make me $100 poorer.
        > > >
        > > > regards,
        > > > Vikram
        > > >
        > > >
        > > >
        > > >
        > > >
        > > > --- In LoadRunner@yahoogroups.com, cheney
        > > > <socheney@> wrote:
        > > >
        > > >
        > > > >
        > > > > Commenting things out and noting that the script
        > > > > "works" requires some definition. Posting
        > > > somethign
        > > > > like this on a group chock full of newbies and
        > > > young
        > > > > eyes is another questionable practice. There's
        > > > > probably room's full of people in Bangalore right
        > > > now
        > > > > commenting things out of their scripts until "it
        > > > > works." --Thanks to your post. :-)
        > > > >
        > > > > Credibility is fine; I give you all the
        > > > credibility
        > > > > in the world; and hope you are proceeding well and
        > > > > according to best practices.
        > > > >
        > > > > But use your credibility responsibly... 1. By
        > > > flaming
        > > > > newbies that don't read the manuals. 2. If you
        > > > > aren't going to flame newbies; and would rather
        > > > attack
        > > > > HP (nee Mercury... ) then there are a huge
        > > > wealth of
        > > > > shortcomings and bugs and bobbles you could go
        > > > after
        > > > > that wouldn't include The siebel web protocol.
        > > > Good
        > > > > lord man!! Why go after siebel web with so much
        > > > > other low hanging fruit around?
        > > > >
        > > > >
        > > > >
        > > > >
        > > > >
        > > > > --- chandna <chandna@> wrote:
        > > > >
        > > > > > Steve,
        > > > > > You are one of acknoweldged expert in
        > > > performance
        > > > > > testing and I am
        > > > > > glad this caught your attention.
        > > > > >
        > > > > > Header to help you avoid reading details: I
        > > > > > implemented all that
        > > > > > Siebel Web does in Web protocol script. Let's
        > > > start
        > > > > > from there. Now
        > > > > > I need reason for why I need to buy Siebel Web.
        > > > > >
        > > > > > I am willing to pay for valuable knowledge and
        > > > > > information and I
        > > > > > appreciate everyone who takes time out to
        > > > respons to
        > > > > > me. And trust
        > > > > > me, I do attempt to research very well. Since I
        > > > am
        > > > > > spend thrift I
        > > > > > did read all of Siebel architecture document
        > > > before
        > > > > > posting this.
        > > > > > I would like to gain some credibility with you
        > > > so
        > > > > > that we do not
        > > > > > focus on personality and focus just on topic. So
        > > > as
        > > > > > example I would
        > > > > > like to share with you that in my college days
        > > > out
        > > > > > of my love for
        > > > > > language I memorized 65K words in English
        > > > dictionary
        > > > > > in a year.
        > > > > > Comes to approximately 18 words per day. I did
        > > > learn
        > > > > > 20 every day
        > > > > > but I had memory leakage of 10%. Also I speak
        > > > five
        > > > > > languages. So I
        > > > > > think I have reasons for existent apart from
        > > > paying
        > > > > > $100. Though
        > > > > > this all doesn't mean I am well rounded in every
        > > > > > fashion as I have
        > > > > > seen myslef making mistakes. Hence the need to
        > > > fly
        > > > > > the kite and I am
        > > > > > also willing to put money where my mouth is.
        > > > > >
        > > > > > Now onto topic of Siebel in detail.
        > > > > > Did you read all of my post?
        > > > > >
        > > > > > 'SWEC comment out' I didn't say. 'SWETS comment
        > > > out'
        > > > > > I didn't say.
        > > > > > I do understand what they stand for.
        > > > > > Rowid I am not leaving hard coded but I noted
        > > > that
        > > > > > even if you leave
        > > > > > it hard coded it does not cause the script to
        > > > fail
        > > > > > and by the logic
        > > > > > I have thought of, it doesn't matter much as far
        > > > as
        > > > > > performance is
        > > > > > concerned. I can explain that later in follow-up
        > > > > > posts.
        > > > > >
        > > > > > I haven't mentioned what all I have done as part
        > > > of
        > > > > > my scripting
        > > > > > SWEC & SWETS included. So probably that's not
        > > > right
        > > > > > approach to
        > > > > > critique. What I am asking for is that someone
        > > > give
        > > > > > me a reason why
        > > > > > should someone need Siebel Web protocol. There
        > > > is no
        > > > > > need to look at
        > > > > > my scripts because whatever Siebel Web does I
        > > > have
        > > > > > learnt that by
        > > > > > recording the script in Siebel Web protocol. And
        > > > > > three additional
        > > > > > things that it does I could have arrived at by
        > > > error
        > > > > > and trial as
        > > > > > well.
        > > > > >
        > > > > > I have done my homework and I have recorded and
        > > > > > tested the scripts
        > > > > > using seven approaches as you find in my earlier
        > > > > > post below. It is
        > > > > > not a matter of conjecture that I deduced that
        > > > > > Siebel Web is doing
        > > > > > nothing special. My pass factor is looking
        > > > closely
        > > > > > at the response
        > > > > > of server including Status message. Also I
        > > > looked at
        > > >
        > > === message truncated ===
        > >
        > >
        > >
        > >
        > >
        > __________________________________________________________
        > _______________
        > > Be a better Globetrotter. Get better travel answers from someone
        > who knows. Yahoo! Answers - Check it out.
        > > http://answers.yahoo.com/dir/?link=list&sid=396545469
        > >
        >
        >
        >
        >
        >
        >
        > ---------------------------------
        > Park yourself in front of a world of choices in alternative
        vehicles.
        > Visit the Yahoo! Auto Green Center.
        >
        > [Non-text portions of this message have been removed]
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.