[topicmapmail] One quibble re: topicmap
Neal McNamara (nealmc)
Those look to be some useful categories. My only quibble would be the
point at which the ontology plugs in. For my type of work, I would
expect it closer to the data & authoring.
Always good to hear from you!
Neal McNamara
-----Original Message-----
From: topicmapma...@infoloom.com
[mailto:topicmapma...@infoloom.com] On Behalf Of
topicmapma...@infoloom.com
Sent: Monday, November 28, 2011 11:00 AM
To: topicm...@infoloom.com
Subject: topicmapmail Digest, Vol 69, Issue 13
Send topicmapmail mailing list submissions to
topicm...@infoloom.com
To subscribe or unsubscribe via the World Wide Web, visit
http://www.infoloom.com/mailman/listinfo/topicmapmail
or, via email, send a message with subject or body 'help' to
topicmapma...@infoloom.com
You can reach the person managing the list at
topicmapm...@infoloom.com
When replying, please edit your Subject line so it is more specific than
"Re: Contents of topicmapmail digest..."
_______________________________________________
topicmapmail mailing list
topicm...@infoloom.com
http://www.infoloom.com/mailman/listinfo/topicmapmail
Patrick Durusau
While creating the image I wondered about where was the "best" place to
put ontology.
What would you suggest as a way to show its possible presence/use in
multiple locations?
One reason for the exercise is that I would like to have several
versions that have a "tool kit" icon for each of the images.
So if you have the clearly labeled "beginner" image, you get different
suggested tools than the "intermediate" (beginner + more) or "expert"
(intermediate + more) or "researcher" (expert + more).
For example, my post yesterday on using Cloud9 to mine Wikipedia in the
cloud is probably an "expert" tool. The post on templates is probably a
research tool.
Hope you are having a great day!
Patrick
On 11/28/2011 05:08 PM, Neal McNamara (nealmc) wrote:
> Hi Patrick,
>
> Those look to be some useful categories. My only quibble would be the
> point at which the ontology plugs in. For my type of work, I would
> expect it closer to the data& authoring.
>
> Always good to hear from you!
>
> Neal McNamara
>
> -----Original Message-----
> From: topicmapma...@infoloom.com
> [mailto:topicmapma...@infoloom.com] On Behalf Of
> topicmapma...@infoloom.com
> Sent: Monday, November 28, 2011 11:00 AM
> To: topicm...@infoloom.com
> Subject: topicmapmail Digest, Vol 69, Issue 13
>
> Send topicmapmail mailing list submissions to
> topicm...@infoloom.com
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://www.infoloom.com/mailman/listinfo/topicmapmail
> or, via email, send a message with subject or body 'help' to
> topicmapma...@infoloom.com
>
> You can reach the person managing the list at
> topicmapm...@infoloom.com
>
> When replying, please edit your Subject line so it is more specific than
> "Re: Contents of topicmapmail digest..."
> _______________________________________________
> topicmapmail mailing list
> topicm...@infoloom.com
> http://www.infoloom.com/mailman/listinfo/topicmapmail
>
--
Patrick Durusau
pat...@durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)
Another Word For It (blog): http://tm.durusau.net
Homepage: http://www.durusau.net
Twitter: patrickDurusau
Alexander Johannesen
An interesting follow-up and possible confirmation is to have various
people with their various systems try to chalk up a similar chart for
their system. Also, I doubt we'll all agree on what is what, even,
like some have ontology and TMCL separate, others have them as joined
efforts (and hence 'ontology' should perhaps be on the other side),
others again use something closer to TMRM which isn't represented, and
so on. For example, my original xSiteable was more of a presentation
layer compiler of Topic Maps websites, and had a very (*very*)
different setup all-together where rendering, storage and delivery was
the same, TMCL (small part) was at the same of of 'ontology', and with
no mining / analyze part between data and the map.
Having said all that, I like it. For those of us who still haven't
given up on total world domination, I think this is a very worthwhile
effort, and would love to see more people chime in. (Might even be a
way to get one leg up RDF and its ilk :)
Regards,
Alex
--
Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps
--- http://shelter.nu/blog/ ----------------------------------------------
------------------ http://www.google.com/profiles/alexander.johannesen ---
Patrick Durusau
On 11/29/2011 09:47 PM, Alexander Johannesen wrote:
> Hiya,
>
> An interesting follow-up and possible confirmation is to have various
> people with their various systems try to chalk up a similar chart for
> their system. Also, I doubt we'll all agree on what is what, even,
> like some have ontology and TMCL separate, others have them as joined
> efforts (and hence 'ontology' should perhaps be on the other side),
> others again use something closer to TMRM which isn't represented, and
> so on. For example, my original xSiteable was more of a presentation
> layer compiler of Topic Maps websites, and had a very (*very*)
> different setup all-together where rendering, storage and delivery was
> the same, TMCL (small part) was at the same of of 'ontology', and with
> no mining / analyze part between data and the map.
>
+1!
Everyone is going to arrange the pieces a little or a lot differently.
My goal was to get a discussion going about those pieces in general,
even if we don't agree all the pieces or where they go.
Knowing where you put the pieces in your application or process is going
to help me explain to someone why to choose your solution over another one.
Or quite honestly, to use good ideas you have for explaining modeling to
users for my own. ;-)
> Having said all that, I like it. For those of us who still haven't
> given up on total world domination, I think this is a very worthwhile
> effort, and would love to see more people chime in. (Might even be a
> way to get one leg up RDF and its ilk :)
>
Thanks!
Actually I think RDF is being held up by its criteria for being
decidable. Few things in human data systems are.
I may have a logic contest some time soon to see how many people use
various forms of logic in the average day.
Suspect the numbers are going to be really low.
If that suspicion is correct, why bother encoding our data into a system
we don't use ourselves?
Hope you are having a great week!
Patrick
Alexander Johannesen
Patrick Durusau <pat...@durusau.net> wrote:
> My goal was to get a discussion going about those pieces in general, even if
> we don't agree all the pieces or where they go.
We're still waiting for people to wake up, I think. :)
> Or quite honestly, to use good ideas you have for explaining modeling to
> users for my own. ;-)
You mean this? ;
https://plus.google.com/u/0/111886865967199209050/posts/QLx3LLeseeD
I have to admit that since doing Topic Maps, the world has for me
become a very different place, and my understanding of the underlying
problems that most software developers struggle with usually come from
places that aren't normally blamed. There really should be an
uprising, but ... well. Coffee.
> I may have a logic contest some time soon to see how many people use various
> forms of logic in the average day. Suspect the numbers are going to be really low.
I take it you mean formal logic? Because if not, the numbers will be
even lower than you suspect, given the wife, and the kids and the
neighbours and my boss and "other people" and ... :)
Regards,
Alex
--
Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps
--- http://shelter.nu/blog/ ----------------------------------------------
------------------ http://www.google.com/profiles/alexander.johannesen ---
Patrick Durusau
On 11/30/2011 11:36 PM, Alexander Johannesen wrote:
> Hiya,
>
> Patrick Durusau<pat...@durusau.net> wrote:
>> My goal was to get a discussion going about those pieces in general, even if
>> we don't agree all the pieces or where they go.
> We're still waiting for people to wake up, I think. :)
>
Maybe yes, maybe no.
I think people are doing mappings between systems all the time, but
informally.
The trick is to find the payoff point where documenting and formalizing
that mapping has some advantage.
Consider the advantages of being opaque:
1) Job security - they always have to ask you for your data
2) Prestige - you are the only one who understands your data
3) Territory - you can convert other data into your format, extending
your range for #1 and #2
I suspect people have at some level woken up and aren't interested in
giving up 1 - 3.
Which is why I am interested in projective topic map techniques, ones
that don't require their cooperation.
>> Or quite honestly, to use good ideas you have for explaining modeling to
>> users for my own. ;-)
> You mean this? ;
> https://plus.google.com/u/0/111886865967199209050/posts/QLx3LLeseeD
Something along those lines, yes. ;-) Perhaps with a few more
walk-through examples.
> I have to admit that since doing Topic Maps, the world has for me
> become a very different place, and my understanding of the underlying
> problems that most software developers struggle with usually come from
> places that aren't normally blamed. There really should be an
> uprising, but ... well. Coffee.
>
;-)
Understand what you mean. Thinking deeply about topic maps will change
your view of many things!
>> I may have a logic contest some time soon to see how many people use various
>> forms of logic in the average day. Suspect the numbers are going to be really low.
> I take it you mean formal logic? Because if not, the numbers will be
> even lower than you suspect, given the wife, and the kids and the
> neighbours and my boss and "other people" and ... :)
>
Well, formal logic is what they are advocating for computer systems.
Not to confuse "formal logic" with "reasoning." Not the same things.
Reasoning can be rough, fuzzy and yet fairly certain, all at the same time.
Reasoning is what we use with subject identity and to think about subjects.
What puzzles me is why we think that taking a subset of that, formal
logic, that will work on computers, is ever going to be as good at some
tasks as we are. Different abilities and so a different outcome.
Hope
> Regards,
>
> Alex
--
Patrick Durusau
pat...@durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)
Another Word For It (blog): http://tm.durusau.net
Homepage: http://www.durusau.net
Twitter: patrickDurusau
Alexander Johannesen
Hiya,
>> We're still waiting for people to wake up, I think. :)
"Patrick Durusau" <pat...@durusau.net> wrote:
>Maybe yes, maybe no.
> I think people are doing mappings between systems all the time, but informally.
Hehe, well I can see maybe my initial response came across a bit pompous. I actually meant, waiting for people to wake up to this thread, since we're the only ones talking ... :)
> I suspect people have at some level woken up and aren't interested in giving up 1 - 3.
If by "people" you mean businesses, then I would agree, but otherwise I've met people thoughout all the world in my travels who understand the modelling argument when brought to them, and they're usually fascinated with topic Maps and / or semantic data modelling as an alternative. I guess if we truly *were* an alternative ..
> Which is why I am interested in projective topic map techniques, ones that don't require their cooperation.
Incidentally, I've decided to write more extensively on this problem of data modelling. I personally think the current practices (and what's being taught in schools) are the biggest cause of all IT sorrows, and needs to be focused on, need to be pointed out the sad state of semantic mismatching even in hetrogenous stacks.
I've lately been giving people the following snippet for the description of a book, and asked if they see anything wrong with it ;
dc:title
foaf:title
It usually hits them a little later after I point out that a duplicate 'title' is *not* the problem, and so after a bit of digging I need to point out that that's two steps away from how easy it should be if it is going to be a success. Most people would agree with that, I think.
>> You mean this? ;
>> https://plus.google.com/u/0/111886865967199209050/posts/QLx3LLeseeD
>
> Something along those lines, yes. ;-) Perhaps with a few more walk-through examples.
I think I need to write another book, or adapt my never-ending manuscript in new directions ...
> What puzzles me is why we think that taking a subset of that, formal logic, that will work on computers, is ever going to be as good at some tasks as we are. Different abilities and so a different outcome.
I've always viewed Topic Maps as a fantastic way to translate between those two worlds in a way that makes sense to both, and people seems to get excited about the prospects of that, especially people who've spent too many years in software development.
Topic Maps is too good for too many things, and just seems to fall into the "too good to be true" bucket before people give it a chance.
Regards,
Alex
Patrick Durusau
On 12/08/2011 06:08 AM, Alexander Johannesen wrote:
Hiya,
>> We're still waiting for people to wake up, I think. :)
"Patrick Durusau" <pat...@durusau.net> wrote:
>Maybe yes, maybe no.
> I think people are doing mappings between systems all the time, but informally.Hehe, well I can see maybe my initial response came across a bit pompous. I actually meant, waiting for people to wake up to this thread, since we're the only ones talking ... :)
The more common "interesting" threads become the more participation you will see in the long run. Noting that different people will join in on different threads, depending upon their time demands and interests.
> I suspect people have at some level woken up and aren't interested in giving up 1 - 3.
If by "people" you mean businesses, then I would agree, but otherwise I've met people thoughout all the world in my travels who understand the modelling argument when brought to them, and they're usually fascinated with topic Maps and / or semantic data modelling as an alternative. I guess if we truly *were* an alternative ..
> Which is why I am interested in projective topic map techniques, ones that don't require their cooperation.
Incidentally, I've decided to write more extensively on this problem of data modelling. I personally think the current practices (and what's being taught in schools) are the biggest cause of all IT sorrows, and needs to be focused on, need to be pointed out the sad state of semantic mismatching even in hetrogenous stacks.
I've lately been giving people the following snippet for the description of a book, and asked if they see anything wrong with it ;
dc:title
foaf:titleIt usually hits them a little later after I point out that a duplicate 'title' is *not* the problem, and so after a bit of digging I need to point out that that's two steps away from how easy it should be if it is going to be a success. Most people would agree with that, I think.
Well, like most things that are interesting, its complicated.
Complicated because the people who developed dc:title don't see much reason to change systems they have built and understand that use dc:* terminology.
With equal cause, the people who have invested in foaf:* feel the same way.
As with all the people who have invested in alternative systems for handling the same information.
Creating yet another system, that over-arches the the existing ones, only creates another system to which there will be adherents.
It is what drives semantic diversity.
What I see as useful about topic maps is that it can accommodate semantic diversity without increasing it. (There is a theoretical caveat to that statement but I forego it in favor of clarity.)
>> You mean this? ;
>> https://plus.google.com/u/0/111886865967199209050/posts/QLx3LLeseeD
>
> Something along those lines, yes. ;-) Perhaps with a few more walk-through examples.I think I need to write another book, or adapt my never-ending manuscript in new directions ...
I keep thinking about using my blog posts as the beginnings of chapters for a book.
> What puzzles me is why we think that taking a subset of that, formal logic, that will work on computers, is ever going to be as good at some tasks as we are. Different abilities and so a different outcome.
I've always viewed Topic Maps as a fantastic way to translate between those two worlds in a way that makes sense to both, and people seems to get excited about the prospects of that, especially people who've spent too many years in software development.
Topic Maps is too good for too many things, and just seems to fall into the "too good to be true" bucket before people give it a chance.
But I also write about software that has the potential to do topic map like processing, whether considered to be topic map engines or no. Because other people when shown an application with capabilities they really want, can then be shown..."there's a topic map inside."
Both are necessary and valuable.
Hope you are having a great week!
Patrick
Regards,
Alex
-- Patrick Durusau pat...@durusau.net Chair, V1 - US TAG to JTC 1/SC 34 Convener, JTC 1/SC 34/WG 3 (Topic Maps) Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300 Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps) OASIS Technical Advisory Board (TAB) - member Another Word For It (blog): http://tm.durusau.net
marijane white
Topic Maps is too good for too many things, and just seems to fall into the "too good to be true" bucket before people give it a chance.
-marijane
Patrick Durusau
What are topic maps about, really?
Topic maps are how we choose to express explanations via computer systems.
It is the taking explanations, "go ask Joe," go ask Alice," etc. and squeezing them into representations in systems that are less capable than a piece of paper.*
When you write a topic map, you are capturing an explanation of how you identify some set of subjects, in hopes that someone else will recognize those subjects.
But since you can't go along with the explanation, topic maps also give you the ability to explain the explanation as it were. To help people recognize the subjects represented by keys.
Topic maps in a very real sense are about capturing how we would explain something, if someone asked us, so they would understand what we meant. We do it all day long in a variety of contexts. It's the writing it down for something far less capable than we are that gets hard.
Hope you are enjoying the new job!
Patrick
*I say computers are less capable than a piece of paper. For example, how many writing directions do you think a computer has? If you said one (1) you would be correct. If you also guessed left to right, you would be correct again. That a computer can display different writing directions is a display capability. It is still reading left to right and humoring you by displaying some other writing direction.
A piece of paper, on the other hand, can start and stop writing in any direction a script can be written.
There are other ways in which a piece of paper exceeds even the coming generation of pentaflop systems. Mark off ten minutes on the clock. How many can you think of?
-marijane
_______________________________________________ topicmapmail mailing list topicm...@infoloom.com http://www.infoloom.com/mailman/listinfo/topicmapmail
-- Patrick Durusau pat...@durusau.net Chair, V1 - US TAG to JTC 1/SC 34 Convener, JTC 1/SC 34/WG 3 (Topic Maps) Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300 Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps) OASIS Technical Advisory Board (TAB) - member Another Word For It (blog): http://tm.durusau.net
Patrick Durusau
Supplement to my prior post:
On 12/08/2011 06:53 PM, Patrick Durusau wrote:
<snip>
> Topic maps in a very real sense are about capturing how we would
> explain something, if someone asked us, so they would understand what
> we meant. We do it all day long in a variety of contexts. It's the
> writing it down for something far less capable than we are that gets
> hard.
>
The other aspect of topic maps is that other people may (or may not)
recognize our explanations (read identifications) as being about the
same subjects as their explanations (read identifications). People do
that all the time.
With topic maps represented in digital form, the options for such
recognition are greatly limited, but still possible.
All topic maps make choices about the explanations/identifications they
can represent.
The history of data representation, aside from topic maps, has been one
of "...everyone can, will, must use *my* system of
explanation/identification...."
There are large ones, like Dublin Core and there are small ones in the
databases that your accounting department has created for their own use.
But none of them encompass all the subjects that could be identified,
not to mention that no system can account for all the subjects that have
prior identifications.
For some purposes, your local accounting department for example, to have
its own internal identification of subjects. It is cost effective, suits
the needs at hand and making it capable of mapping to other
identifications has no value. The larger organization's accounting
structure might disagree but that would be at their expense.
Hope you are having a great day!
Patrick
--
marijane white
-marijane