CIS/2 workshop documents
26 views
Skip to first unread message
Robert Lipman
May 26, 2010, 12:53:35 PM5/26/10
to CIS/2 - CIMsteel Integration Standards
Hi,
In the Files section, I have uploaded two documents related to the CIS/
2 workshop on May 11 at NASCC in Orlando.
The first is a few slides used during the workshop
http://cis2-group.googlegroups.com/web/CIS2_Workshop_NASCC_2010.pdf
and the second is some of the results from the brainstorming session.
http://cis2-group.googlegroups.com/web/cis2_workshop_2010.pdf
---
Robert Lipman
http://cic.nist.gov/lipman/
TwH
May 29, 2010, 9:09:24 AM5/29/10
to CIS/2 - CIMsteel Integration Standards
The uploaded pdf's seem to show a few chinks in the armor of CIS/2 in
that the data format is not universally implemented, not universally
translatable and not universally compatible for data exchanges. No
software vendor would really want that, would they?
From my study, the real "value add" of CIS/2 is creating a downstream
path from an engineering design model (CIS/2-A) through a detailing
model (CIS/2-B) to a fabrication model (CIS/2-C). Another path is
from CIS/2(A,B) translation to an IFC model for coordination efforts
(e.g. Navisworks clash detection). Unfortunately, these are generally
downstream trips. Again, the software vendors have no real incentive
to move the format back upstream, perhaps enabling a competitor
product to import a model file. So the CIS/2 focus should be how to
streamline design-to-manufacturing. Some of my acquaintances suggest
SDNF format is a bit more robust than CIS/2 for data integration
purposes, but I don't know that for certain. In any case, let's
continue to solve the value proposition as well as practical!
that the data format is not universally implemented, not universally
translatable and not universally compatible for data exchanges. No
software vendor would really want that, would they?
From my study, the real "value add" of CIS/2 is creating a downstream
path from an engineering design model (CIS/2-A) through a detailing
model (CIS/2-B) to a fabrication model (CIS/2-C). Another path is
from CIS/2(A,B) translation to an IFC model for coordination efforts
(e.g. Navisworks clash detection). Unfortunately, these are generally
downstream trips. Again, the software vendors have no real incentive
to move the format back upstream, perhaps enabling a competitor
product to import a model file. So the CIS/2 focus should be how to
streamline design-to-manufacturing. Some of my acquaintances suggest
SDNF format is a bit more robust than CIS/2 for data integration
purposes, but I don't know that for certain. In any case, let's
continue to solve the value proposition as well as practical!
Mark
Jun 3, 2010, 6:06:04 AM6/3/10
to CIS/2 - CIMsteel Integration Standards
Regarding the comment about the SDNF file is more robust compared to
the CIS/2 file I have the following two comments.
First as far back as the mid 90’s it was recognized that the
fundamental problem the software developer faced was the issue of
competing standards. For this reason AISC took on the task of trying
to select a single standard that would fulfill the needs of the steel
industry’s data exchange requirements. A team was recruited that
included everyone form software develop, detailers, engineers,
fabricators, etc… The CIS/2 standard was compared to the version of
the SDNF file we had at the time and was not found to have the same
ability to deal with exchanges needs. Later a 3.0 version was
developed. It too lacked the same ability to deal with the overall
needs identified in the CIS/2 standard. Because of the complete nature
of the CIS/2 standard compared to the SDNF files both new and old it
is more difficult to implement which means that if the less complex
SDNF file is used you don’t get everything someone in the supply chain
will need.
The second and more important is the idea of competing standards. If
we as an industry request the software developers to support more than
one standard we will never achieve the goal of interoperability we so
desperately need. Right now we have a 10 plus year effort to develop
the CIS/2 standard. We have some success stories using this standard
but since all the application I have support more that one standard we
can’t drive the software developers to put all there efforts against
one steel standard. Once we leave the steel supply chain we are now
dealing with the IFC format.
The work at Georgia Tech is helping use see that there are exchanges
that are best done using CIS/2 while other are best done using the IFC
for mate. To bring the less complex SDNF format back into the equation
will only dilute the work done to date.
> > Robert Lipmanhttp://cic.nist.gov/lipman/- Hide quoted text -
>
> - Show quoted text -
the CIS/2 file I have the following two comments.
First as far back as the mid 90’s it was recognized that the
fundamental problem the software developer faced was the issue of
competing standards. For this reason AISC took on the task of trying
to select a single standard that would fulfill the needs of the steel
industry’s data exchange requirements. A team was recruited that
included everyone form software develop, detailers, engineers,
fabricators, etc… The CIS/2 standard was compared to the version of
the SDNF file we had at the time and was not found to have the same
ability to deal with exchanges needs. Later a 3.0 version was
developed. It too lacked the same ability to deal with the overall
needs identified in the CIS/2 standard. Because of the complete nature
of the CIS/2 standard compared to the SDNF files both new and old it
is more difficult to implement which means that if the less complex
SDNF file is used you don’t get everything someone in the supply chain
will need.
The second and more important is the idea of competing standards. If
we as an industry request the software developers to support more than
one standard we will never achieve the goal of interoperability we so
desperately need. Right now we have a 10 plus year effort to develop
the CIS/2 standard. We have some success stories using this standard
but since all the application I have support more that one standard we
can’t drive the software developers to put all there efforts against
one steel standard. Once we leave the steel supply chain we are now
dealing with the IFC format.
The work at Georgia Tech is helping use see that there are exchanges
that are best done using CIS/2 while other are best done using the IFC
for mate. To bring the less complex SDNF format back into the equation
will only dilute the work done to date.
>
> - Show quoted text -
TwH
Jun 3, 2010, 1:03:14 PM6/3/10
to CIS/2 - CIMsteel Integration Standards
Mark, thanks for your comments. I appreciate what you're saying, and
generally agree with AISC championing CIS/2. I personally believe
that there will never be a data exchange "standard" for structural
steel even though CIS/2 is a very logical front-runner. Software
vendors generally have no incentive to create a common exchange format
since it opens the doors for competitors to access data created in a
(their) proprietary format. And the steel design (engineering)
software industry is too fractured to create a standard that can flow
downstream to detailing and there is the risk of data-sharing that
attorneys and insurance companies have identified to further reduce
the incentive for data-sharing solutions. So my framework is to allow
a multiplicity of data formats downstream all held within a common
repository that is wrapped with a legal contract. Call it IPD meets
BIM. My name for this is "CIS/3D". Don't you think that if engineers
and detailsers/fabricators worked together with one progressive
engineering model and derivative detailing models created "on the fly"
from the engineering model we truly could collaborate? Who cares if
it's CIS/2 or SDNF or IFC if the next user can use it productively?
Let me know what you think. Some friends and I are building this new
framework for collaboration currently.
> > > Robert Lipmanhttp://cic.nist.gov/lipman/-Hide quoted text -
generally agree with AISC championing CIS/2. I personally believe
that there will never be a data exchange "standard" for structural
steel even though CIS/2 is a very logical front-runner. Software
vendors generally have no incentive to create a common exchange format
since it opens the doors for competitors to access data created in a
(their) proprietary format. And the steel design (engineering)
software industry is too fractured to create a standard that can flow
downstream to detailing and there is the risk of data-sharing that
attorneys and insurance companies have identified to further reduce
the incentive for data-sharing solutions. So my framework is to allow
a multiplicity of data formats downstream all held within a common
repository that is wrapped with a legal contract. Call it IPD meets
BIM. My name for this is "CIS/3D". Don't you think that if engineers
and detailsers/fabricators worked together with one progressive
engineering model and derivative detailing models created "on the fly"
from the engineering model we truly could collaborate? Who cares if
it's CIS/2 or SDNF or IFC if the next user can use it productively?
Let me know what you think. Some friends and I are building this new
framework for collaboration currently.
DDDE
Jun 4, 2010, 10:18:51 AM6/4/10
to CIS/2 - CIMsteel Integration Standards
I am curious to see your framework. Anything you can share with me?
Doug
> > > > Robert Lipmanhttp://cic.nist.gov/lipman/-Hidequoted text -
Doug
Ivan jivkov
Jun 4, 2010, 4:06:30 PM6/4/10
to CIS/2 - CIMsteel Integration Standards
This is a very interesting concept. It makes a lot of sense. I will be
curious to know how rich is the data in that CIS/3D. How far beyond
detailing can we go with the data?
curious to know how rich is the data in that CIS/3D. How far beyond
detailing can we go with the data?
Robert Lipman
Jun 4, 2010, 4:16:13 PM6/4/10
to CIS/2 - CIMsteel Integration Standards
Tom,
Is your "CIS/3D" the same/similar/different/better than what you
showed me several years ago with the product and concepts for you
company Saint-Joe?
Is your "CIS/3D" the same/similar/different/better than what you
showed me several years ago with the product and concepts for you
company Saint-Joe?
TwH
Jun 6, 2010, 10:42:17 AM6/6/10
to CIS/2 - CIMsteel Integration Standards
CIS/3D is the name for the collaboration servers & process. See the
post to Doug at SDS/2 for the full description of the working parts of
the system.
post to Doug at SDS/2 for the full description of the working parts of
the system.
TwH
Jun 6, 2010, 10:39:59 AM6/6/10
to CIS/2 - CIMsteel Integration Standards
Doug, an IT friend (Garrett Burton) and I have filed for a U.S. patent
(application 12/082,479 on April 11, 2008 for a "design collaboration
system.") You can search for this at USTPO.gov to get the
particulars, but here is the general concept. This patent is a
business method based upon a mix of hardware and software that will
directly replicate data files in near-real time between collaborators,
primarily focusing on the steel construction indusry.
The outcome of this technology is "increasing data flow" between
partners within a secure collaboration & data sharing agreement.
As an example of a use case, a structural engineer can create and
update an analytical model, say a RAMsteel, STAAD file, CIS/2 etc.
When posted or updated to the engineer's local server, it is then
"shared" to other collaborators inside the "trusted space" of the
network mirroring servers. Since the data is protected by a data-
sharing license agreement between users, it creates a "collaborative
data space with an IPD wrapper." This could potentially overcome the
current construction industry problems of working together and sharing
data files. Then for example, after the engineering model is
uploaded, a steel detailer can load the CIS/2 and quickly create an
SDS/2 model and pricing ABM from the latest model, eliminating a great
deal of manual take-off time. In some cases, the detailing model
could also flow back to the designer (e.g. an IFC or Revit model from
the detailing software), which can then become either an updated
analytical or reference design model. Another outcome of using this
method is to allow concurrent detailing using one model which is
shared through mirroring replication, so each site is up to date in
near real time. This method also can provide continual off-site
backup of critical data. The mirroring servers features Linux scripts
to 1) mount into a Windows environment as NAS-SAN drives, 2) provide
integral version control and roll-back and 3) provide automated data
management between collaborators. Each feature provides direct
productivity benefits since the users no longer have to spend as much
time on data management and can spend more on modeling and project
outcomes. Anyway, this is a bit of how it works from about 30,000
feet. Let me know what issues, flaws or problems you see after you
think about this a bit.
Philosophically, I'd call it a sort of a "middleware" -- hardware and
software bundled to enable rapid data-sharing collaboration in a
trusted (i.e. contractual) space.
Apologies for the length of reply.
> > > > > Robert Lipmanhttp://cic.nist.gov/lipman/-Hidequotedtext -
(application 12/082,479 on April 11, 2008 for a "design collaboration
system.") You can search for this at USTPO.gov to get the
particulars, but here is the general concept. This patent is a
business method based upon a mix of hardware and software that will
directly replicate data files in near-real time between collaborators,
primarily focusing on the steel construction indusry.
The outcome of this technology is "increasing data flow" between
partners within a secure collaboration & data sharing agreement.
As an example of a use case, a structural engineer can create and
update an analytical model, say a RAMsteel, STAAD file, CIS/2 etc.
When posted or updated to the engineer's local server, it is then
"shared" to other collaborators inside the "trusted space" of the
network mirroring servers. Since the data is protected by a data-
sharing license agreement between users, it creates a "collaborative
data space with an IPD wrapper." This could potentially overcome the
current construction industry problems of working together and sharing
data files. Then for example, after the engineering model is
uploaded, a steel detailer can load the CIS/2 and quickly create an
SDS/2 model and pricing ABM from the latest model, eliminating a great
deal of manual take-off time. In some cases, the detailing model
could also flow back to the designer (e.g. an IFC or Revit model from
the detailing software), which can then become either an updated
analytical or reference design model. Another outcome of using this
method is to allow concurrent detailing using one model which is
shared through mirroring replication, so each site is up to date in
near real time. This method also can provide continual off-site
backup of critical data. The mirroring servers features Linux scripts
to 1) mount into a Windows environment as NAS-SAN drives, 2) provide
integral version control and roll-back and 3) provide automated data
management between collaborators. Each feature provides direct
productivity benefits since the users no longer have to spend as much
time on data management and can spend more on modeling and project
outcomes. Anyway, this is a bit of how it works from about 30,000
feet. Let me know what issues, flaws or problems you see after you
think about this a bit.
Philosophically, I'd call it a sort of a "middleware" -- hardware and
software bundled to enable rapid data-sharing collaboration in a
trusted (i.e. contractual) space.
Apologies for the length of reply.
Robert Lipman
Jun 7, 2010, 9:30:18 AM6/7/10
to CIS/2 - CIMsteel Integration Standards
Tom,
Could you provide a direct link to the patent application. I could
not find it on the USTPO website.
Could you provide a direct link to the patent application. I could
not find it on the USTPO website.
TwH
Jun 8, 2010, 10:22:18 AM6/8/10
to CIS/2 - CIMsteel Integration Standards
http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.html&r=2&f=G&l=50&d=PG01&p=1&S1=%28hartmann.IN.+AND+collaboration%29&OS=in/hartmann+AND+collaboration&RS=%28IN/hartmann+AND+collaboration%29
It is in the patent application, not yet granted. If you search for
hartmann as the inventor and collaboration as a word, it will show
up. The above link is a bit unwieldy.
It is in the patent application, not yet granted. If you search for
hartmann as the inventor and collaboration as a word, it will show
up. The above link is a bit unwieldy.
Reply all
Reply to author
Forward
0 new messages