|
Am 08.04.2026 um 16:30 schrieb Jim Oberholtzer <midrangel@xxxxxxxxxxxxxxxxx>:
If you are requesting IBM to integrate the BOB functions into RDi, you might be waiting a very long time for that. IBM chooses to put that functionality in VSCode and that is the direction they are going. That was telegraphed loud and clear on several occasions, for instance Merlin (like it or not) was delivered in VSCode. Now BOB. Clue phone is ringing folks. I know that reading IBM tea leaves is sometimes an art, but in this case it’s somewhat obvious.
Forta may choose to put that functionality into RDi but I would seriously doubt that it will, even if contractually they can.
Jim Oberholtzer
Agile Technology Architects
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> on behalf of cesco via MIDRANGE-L <midrange-l@xxxxxxxxxxxxxxxxxx>
Date: Wednesday, April 8, 2026 at 9:21 AM
To: cesco via MIDRANGE-L <midrange-l@xxxxxxxxxxxxxxxxxx>
Cc: cesco <emaxt6@xxxxxxxxx>
Subject: Re: VSCode and Bob
Hello Daniel, thanks for the update.
I have some copies at closing day time in git to just track changes and query, but the live development is on members.
I'll wait till the premium package is announced.
I would personally also like to have something in RDi (like an "AI outline" that query one of the LLM) right there instead that reinventing the wheel.
For the time being I'm experimenting with the IBMi MCP server , and I query it via a known payed LLM with a well knoen good reasoning model that also integrate all the various online knowledge.
To let the LLM retrieve the source directly in a conversation while avoiding it to let create things to avoid risk (ie. allow just SQL select...) I published a UDTF to return a source member via SQL (I don't know if there is something already stock) to let the LLM pick the sources it wants to analyze.
I share here in case somebody find it useful or are there alternatives/comments
CREATE OR REPLACE FUNCTION MYLIB.GET_SOURCE_MEMBER_T (
P_LIBRARY VARCHAR(10),
P_FILE VARCHAR(10),
P_MEMBER VARCHAR(10)
)
RETURNS TABLE (
SRCSEQ DECIMAL(6, 2),
SRCDAT DECIMAL(6, 0),
SRCDTA VARCHAR(240)
)
LANGUAGE SQL
SPECIFIC MYLIB/GETSRCMBRT
MODIFIES SQL DATA
NOT DETERMINISTIC
NO EXTERNAL ACTION
DISALLOW PARALLEL
SET OPTION COMMIT = *NONE, USRPRF = *USER, DYNUSRPRF = *USER
BEGIN
DECLARE V_ALIAS VARCHAR(100);
DECLARE V_STMT VARCHAR(500);
DECLARE V_SEQ DECIMAL(6, 2);
DECLARE V_DAT DECIMAL(6, 0);
DECLARE V_DTA VARCHAR(240);
DECLARE V_AT_END INT DEFAULT 0;
DECLARE C1 CURSOR FOR S1;
DECLARE CONTINUE HANDLER FOR SQLSTATE '02000' SET V_AT_END = 1;
-- unique alias name to access the member
SET V_ALIAS = 'GETSRCMBRT' || GENERATE_UUID();
-- Point the alias at the requested source member
SET V_STMT = 'CREATE ALIAS QTEMP."' || V_ALIAS || '" FOR "' ||
P_LIBRARY || '"."' || P_FILE || '"("' || P_MEMBER || '")';
EXECUTE IMMEDIATE V_STMT;
-- Prepare a dynamic SELECT against that alias
SET V_STMT = 'SELECT SRCSEQ, SRCDAT, CAST(SRCDTA AS VARCHAR(240)) ' ||
'FROM QTEMP."' || V_ALIAS || '"';
PREPARE S1 FROM V_STMT;
OPEN C1;
FETCH FROM C1 INTO V_SEQ, V_DAT, V_DTA;
WHILE V_AT_END = 0 DO
PIPE (V_SEQ, V_DAT, V_DTA);
FETCH FROM C1 INTO V_SEQ, V_DAT, V_DTA;
END WHILE;
CLOSE C1;
-- Clean up QTEMP
SET V_STMT = 'DROP ALIAS QTEMP."' || V_ALIAS || '"';
EXECUTE IMMEDIATE V_STMT;
RETURN;
END
On Tuesday, April 7, 2026 at 03:26:38 PM GMT+2, cesco via MIDRANGE-L <midrange-l@xxxxxxxxxxxxxxxxxx> wrote:
I have still to try Bob, do you confirm that at the present release cannot work directly with classical source members directly (like i.e. RDi), in order to quickly introduce it directly on the on system dev (compile from source member, generate member DSPF directly etc.)?
by the way, I started using the IBMi MCP from IBM to integrate the tooling into my LLM to query the IBMi system, works fine, maybe some tooling should be better parametrized, but it is still version < 1 so not much to complain, and it is good to have a stock MCP without reinventing the wheel.
https://github.com/IBM/ibmi-mcp-server
Do anybody know if there is a commitment from IBM for this MCP or is considered experimental?
thx
On Tuesday, April 7, 2026 at 01:32:53 AM GMT+2, Vern Hamberg via MIDRANGE-L <midrange-l@xxxxxxxxxxxxxxxxxx> wrote:
To amplify what Justin posted a bit, you can get all the VS Code
extensions in Bob as you do now with VS Code. So you can do things with
other AI tools as you do now, with the plus (we hope) of IBM-tailored
functionality.
*Regards*
*Vern Hamberg*
IBM Champion 2025 <cid:part1.uYM0Dee4.bQcfQTNi@centurylink.net> CAAC
(COMMON Americas Advisory Council) IBM Influencer 2023
On 4/6/2026 8:13 AM, eric bothes wrote:--
i've been doing this for a while and have seen ibm push a product than
its does away. i am not a big fan of the way BOB is being presented. I
would have preferred a true VSC addon rather than a replacement. I
have many other projects i use VSC for and having "another" IDE does
not work well.
In addition, not a big fan of BOB COINS. The get used up rather
quickly even if the code generated was "incorrect" and BOB says "oh,
yeah. i will fix that". I found better results using other AI engines.
I had big hope for BOB, but the ibm history of products leaves we weary.
On 4/6/2026 8:47 AM, Rob Berendt wrote:
Is this the list for such discussions? Or the WDSC list? Or is WDSC
pretty much just for RDi?
What I am wondering about is if IBM Bob is not an add on to VSCode but
basically an integrated package including VSCode is if the VSCode may
lag
behind. Compare this to RDi and how current it stays with levels of
Eclipse.
Not an original thought on my part. Someone else asked me, so inquiring
minds want to know.
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2026 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.
midrange.com
