IDS vs MVD for IFC checking

[andyward]

Warning: possible confrontational (3rd party) opinion ahead:

Supposedly, according to LinkedIn, IDS is 'useless' and we should all be using MVDs /MvdXml for verifying IFC against client requirements because that's what is defined in ISO16739.:2024 for checking information exchanges. IDS has no part to play in model checking, and all its advocates are misleading the openBIM industry. I'm paraphrasing, but that's the gist of a long thread on LinkedIn (which I won't link to here to avoid a pile on). This was from a 'respected' openBIM / COBie advocate.

I wanted to get a sense-check on why buildingSMART may have changed their views on MVD for verification since the ISO16739 publication. e.g. these articles potentially contradict the ISO

https://www.buildingsmart.org/information-requirements/
https://technical.buildingsmart.org/standards/ifc/mvd/

Obviously these aren't as authoritative as an ISO standard, but they promote IDS as the right choice for end-user validation while MVDs are standards that need to be implemented by software vendors. All of which has always made sense to me. I can count custom MVDs supplied by clients on the fingers of one hand.

So if this guidance is in conflict with the ISO clauses what takes precedence? Any thoughts?

[atomczak]

In the introduction to IFC 4.3, it says there are only 3 official MVDs for IFC 4.3, and they are to be understood as levels of implementation for software to be certified against. https://ifc43-docs.standards.buildingsmart.org/IFC/RELEASE/IFC4x3/HTML/content/introduction.htm#Model-View-Definitions

In the latest buildingSMART technical roadmap, it is explained in detail why MVD is not suitable for the project-specific information requirements and why IDS was proposed instead: https://www.buildingsmart.org/wp-content/uploads/2020/09/20200430_buildingSMART_Technical_Roadmap.pdf
Another article providing a good explanation of the problems of MVD: https://www.buildingsmart.org/the-curious-case-of-the-mvd/

[andyward]

Cheers @atomczak useful links - especially section 3.4 of the Roadmap which I'd not seen previously.

I guess until some of this is codified into a future ISO we're just going to have a few traditionalists telling us we're doing this all wrong.

[aothms]

Thanks for bringing this up Andy. Trying to stay factual: "... because that's what is defined in ISO16739.:2024". Are there references to mvdxml there? Is there any actual text that disqualifies IDS? Sent from a mobile device, excuse my brevity. Kind regards, Thomas Op wo 26 nov 2025, 18:16 schreef Andy Ward ***@***.***>:

…

Cheers @atomczak <https://github.com/atomczak> useful links - especially section 3.4 of the Roadmap which I'd not seen previously. I guess until some of this is codified into a future ISO we're just going to have a few traditionalists telling us we're doing this all wrong. — Reply to this email directly, view it on GitHub <#439 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAILWVZ6Y43NSZC4H4NQU7T36XN5BAVCNFSM6AAAAACNIXA6WWVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKMBYHA4DQOA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[andyward]

Good question. I can't tell for sure as I only have a 2018 version at hand. I have probably inferred the MvdXml standard from various statements in a long thread stating MVD should be used in place of IDS. I'll save you the lengthy diatribe, but this was one such statement:

models need only to be tested against their MVD schemas. Both software developers and practitioners. That’s one of the main purposes of MVDs. ...

[billeast]

On the list of IFC 4.3 MVD's should also be the approved, but not yet implemented/tested, AOH-BSEM (replacing US COBie 2.4) and associated AOH Framework (for any other project type/use case handover data set.) Get Outlook for Android<https://aka.ms/AAb9ysg>

…

________________________________ From: Andy Ward ***@***.***> Sent: Wednesday, November 26, 2025 12:04:53 PM To: buildingSMART/IDS ***@***.***> Cc: Subscribed ***@***.***> Subject: Re: [buildingSMART/IDS] IDS vs MVD for IFC checking (Discussion #439) Good question. I can't tell for sure as I only have a 2018 version at hand. I have probably inferred the MvdXml standard from various statements in a long thread stating MVD should be used in place of IDS. I'll save you the lengthy diatribe, but this was one such statement: models need only to be tested against their MVD schemas. Both software developers and practitioners. That’s one of the main purposes of MVDs. ... — Reply to this email directly, view it on GitHub<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_buildingSMART_IDS_discussions_439-23discussioncomment-2D15089335&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=V_aOf878tBKvVp9NMtdjAkoq3L_u8C9cfNv-stcPu5XxDS5kMAdzmwZVC-UKGgRb&m=JohOaIyTPVPgWIYvaFOaTE8IsYQgczt6u-eyMlSLFgRu_q0xSL-IooR_bFNlcNDq&s=v80rleoywSIJF9dj7FFV6Uycl9k960s5sNrTyZY-nPI&e=>, or unsubscribe<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ABTDCBIH4N5DWBAHSXKCJLL36XTULAVCNFSM6AAAAACNIXA6WWVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKMBYHEZTGNI&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=V_aOf878tBKvVp9NMtdjAkoq3L_u8C9cfNv-stcPu5XxDS5kMAdzmwZVC-UKGgRb&m=JohOaIyTPVPgWIYvaFOaTE8IsYQgczt6u-eyMlSLFgRu_q0xSL-IooR_bFNlcNDq&s=gE3wt99xofb1AhE1dG4f7cfpNxJe5LbVhV9voyGbp8I&e=>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>