Exploring ManifoldCF ramifications
Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl
Re: Exploring ManifoldCF ramifications
So, clearly, we need better names to work with, since the original list has led to another impass. Look at http://thesaurus.com/browse/multiple for some thesaurus equivalents of multiple or manifold. Some interesting ones: aggregate conglomerate legion diversiform multifold The first two are not specific enough, in my opinion, but the last three have possibilities. Looking around on the web: multifold is apparently open diversiform is an online jewelry store legion has occasional references to a simulation engine, but is otherwise unused I could withdraw the vote, and call a vote instead for Apache Multifold or Apache Legion, if there's enough buy-in for these possibilities. Karl On Tue, Sep 21, 2010 at 9:28 AM, Karl Wright daddy...@gmail.com wrote: Do you wish to change your vote, in that case? Karl On Tue, Sep 21, 2010 at 8:02 AM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: I'd much prefer a simple, short, name. Using a descriptive phrase as a name has these problems. Tacking on CF does indeed fix one problem, but at a high cost. That said, I am okay with a combo of a short name and a long name. So, if the short name were Ralph, the long name would be Apache Ralph Connectors Framework and we would speak of either the Apache Ralph Connectors Framework or just Ralph. Class names would begin with the capitalized short name, Ralph, and package and file names would use the lower-case, ralph as in org.apache.ralph.core.interfaces.RalphException. And upon graduation, the project would be housed at http://ralph.apache.org/. Now, I wasn't seriously considering Ralph as a name for LCF, but... it works for me. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 3:16 AM To: connectors-dev connectors-dev@incubator.apache.org Subject: Exploring ManifoldCF ramifications Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl
Re: Exploring ManifoldCF ramifications
The exception naming issue is noted, but that's really a separate problem. The IOException exception comes from the IO subsystem, and it's the base exception of everything from an encoding exception through a socket problem through a timeout. ACFException is a similar base exception class, except it comes from ACF. So there is a rough parity there. If you want to challenge the use of base exception classes, so be it, but that's not the difficulty with ManifoldCF. Maybe we don't understand your intended usage of ManifoldCF, since it seems to me like you possibly meant Apache Manifold Connectors Framework for the full name? If so, I certainly don't think any of us got that. Can you clarify/confirm? Karl On Tue, Sep 21, 2010 at 9:58 AM, Grant Ingersoll gsing...@apache.org wrote: Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
I concur with having a base class for all ACF exceptions, and then the specific exceptions extend that base. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 10:08 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications The exception naming issue is noted, but that's really a separate problem. The IOException exception comes from the IO subsystem, and it's the base exception of everything from an encoding exception through a socket problem through a timeout. ACFException is a similar base exception class, except it comes from ACF. So there is a rough parity there. If you want to challenge the use of base exception classes, so be it, but that's not the difficulty with ManifoldCF. Maybe we don't understand your intended usage of ManifoldCF, since it seems to me like you possibly meant Apache Manifold Connectors Framework for the full name? If so, I certainly don't think any of us got that. Can you clarify/confirm? Karl On Tue, Sep 21, 2010 at 9:58 AM, Grant Ingersoll gsing...@apache.org wrote: Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
My interpretation from the beginning is that there is a formal name prefixed with Apache that would get used external to the project to refer to the project, but then within the project we would just use the shorthand name, whether that means simply dropping the Apache or abbreviating the name with an acronym. If the project name was a short name to begin with, then abbreviation would not be needed, but if the name is too long and clumsy, an abbreviation might be called for. Manifold would fit the short prescription fine, but with ManifoldCF, the temptation to shorten it (some people, like me, are clumsy with too much shift key action) to MCF is somewhat... obvious. And when you lower-case the name for package names to manifoldcf, it kind of looks weird. -- Jack Krupansky -- From: Grant Ingersoll gsing...@apache.org Sent: Tuesday, September 21, 2010 9:58 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
On 9/21/10 10:08 AM, Karl Wright wrote: The exception naming issue is noted, but that's really a separate problem. The IOException exception comes from the IO subsystem, and it's the base exception of everything from an encoding exception through a socket problem through a timeout. ACFException is a similar base exception class, except it comes from ACF. So there is a rough parity there. If you want to challenge the use of base exception classes, so be it, but that's not the difficulty with ManifoldCF. You're not on your own here - see the heavily used SolrException from Solr ;)
Re: Exploring ManifoldCF ramifications
Butting in here. You can 'twist' the manifold word in other ways, e.g. manifolio, or some such - full name The Apache Manifolio Connector Framework, short name manifolio. Upayavira On Tue, 2010-09-21 at 10:26 -0400, Jack Krupansky wrote: My interpretation from the beginning is that there is a formal name prefixed with Apache that would get used external to the project to refer to the project, but then within the project we would just use the shorthand name, whether that means simply dropping the Apache or abbreviating the name with an acronym. If the project name was a short name to begin with, then abbreviation would not be needed, but if the name is too long and clumsy, an abbreviation might be called for. Manifold would fit the short prescription fine, but with ManifoldCF, the temptation to shorten it (some people, like me, are clumsy with too much shift key action) to MCF is somewhat... obvious. And when you lower-case the name for package names to manifoldcf, it kind of looks weird. -- Jack Krupansky -- From: Grant Ingersoll gsing...@apache.org Sent: Tuesday, September 21, 2010 9:58 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
That's a perfect example of what I was trying to suggest and avoids the usage problems. Although it has too many syllables for my taste, but that's just me. -- Jack Krupansky -- From: Upayavira u...@odoko.co.uk Sent: Tuesday, September 21, 2010 10:39 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Butting in here. You can 'twist' the manifold word in other ways, e.g. manifolio, or some such - full name The Apache Manifolio Connector Framework, short name manifolio. Upayavira On Tue, 2010-09-21 at 10:26 -0400, Jack Krupansky wrote: My interpretation from the beginning is that there is a formal name prefixed with Apache that would get used external to the project to refer to the project, but then within the project we would just use the shorthand name, whether that means simply dropping the Apache or abbreviating the name with an acronym. If the project name was a short name to begin with, then abbreviation would not be needed, but if the name is too long and clumsy, an abbreviation might be called for. Manifold would fit the short prescription fine, but with ManifoldCF, the temptation to shorten it (some people, like me, are clumsy with too much shift key action) to MCF is somewhat... obvious. And when you lower-case the name for package names to manifoldcf, it kind of looks weird. -- Jack Krupansky -- From: Grant Ingersoll gsing...@apache.org Sent: Tuesday, September 21, 2010 9:58 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
Multifold is a real word but is rarely used, which is why it seemed to solve all the problems too. Karl On Tue, Sep 21, 2010 at 10:49 AM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: That's a perfect example of what I was trying to suggest and avoids the usage problems. Although it has too many syllables for my taste, but that's just me. -- Jack Krupansky -- From: Upayavira u...@odoko.co.uk Sent: Tuesday, September 21, 2010 10:39 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Butting in here. You can 'twist' the manifold word in other ways, e.g. manifolio, or some such - full name The Apache Manifolio Connector Framework, short name manifolio. Upayavira On Tue, 2010-09-21 at 10:26 -0400, Jack Krupansky wrote: My interpretation from the beginning is that there is a formal name prefixed with Apache that would get used external to the project to refer to the project, but then within the project we would just use the shorthand name, whether that means simply dropping the Apache or abbreviating the name with an acronym. If the project name was a short name to begin with, then abbreviation would not be needed, but if the name is too long and clumsy, an abbreviation might be called for. Manifold would fit the short prescription fine, but with ManifoldCF, the temptation to shorten it (some people, like me, are clumsy with too much shift key action) to MCF is somewhat... obvious. And when you lower-case the name for package names to manifoldcf, it kind of looks weird. -- Jack Krupansky -- From: Grant Ingersoll gsing...@apache.org Sent: Tuesday, September 21, 2010 9:58 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
If you are looking for non-existent words built out of parts that nevertheless have meanings, try these: manilink, manicon, maniweb, manisource multicon, multiweb heterolink, heteroweb, heterosource Karl On Tue, Sep 21, 2010 at 11:00 AM, Karl Wright daddy...@gmail.com wrote: Multifold is a real word but is rarely used, which is why it seemed to solve all the problems too. Karl On Tue, Sep 21, 2010 at 10:49 AM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: That's a perfect example of what I was trying to suggest and avoids the usage problems. Although it has too many syllables for my taste, but that's just me. -- Jack Krupansky -- From: Upayavira u...@odoko.co.uk Sent: Tuesday, September 21, 2010 10:39 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Butting in here. You can 'twist' the manifold word in other ways, e.g. manifolio, or some such - full name The Apache Manifolio Connector Framework, short name manifolio. Upayavira On Tue, 2010-09-21 at 10:26 -0400, Jack Krupansky wrote: My interpretation from the beginning is that there is a formal name prefixed with Apache that would get used external to the project to refer to the project, but then within the project we would just use the shorthand name, whether that means simply dropping the Apache or abbreviating the name with an acronym. If the project name was a short name to begin with, then abbreviation would not be needed, but if the name is too long and clumsy, an abbreviation might be called for. Manifold would fit the short prescription fine, but with ManifoldCF, the temptation to shorten it (some people, like me, are clumsy with too much shift key action) to MCF is somewhat... obvious. And when you lower-case the name for package names to manifoldcf, it kind of looks weird. -- Jack Krupansky -- From: Grant Ingersoll gsing...@apache.org Sent: Tuesday, September 21, 2010 9:58 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Let's not overly analyze things here. I'm not saying we need to pick Manifold CF, but if we do, we certainly can solve these writing issues by either re-writing the sentences in question (instead of search/replace) and just use MCF. As for the Exceptions, I find an exception named ACFException meaningless to an app dev. anyway. Duh it's an ACFException, it came from ACF. You don't call an IOException a JavaException just b/c it came from Java, you give it a name that relates to the thing that went wrong, as in something went wrong doing IO. Give it a name that says what happened. On Sep 21, 2010, at 3:16 AM, Karl Wright wrote: Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl -- Grant Ingersoll http://lucenerevolution.org Apache Lucene/Solr Conference, Boston Oct 7-8
Re: Exploring ManifoldCF ramifications
It sounds like the vote is moot pending resolution of the doc/code usage issue, but I'll go ahead and withdraw my vote at least temporarily pending further evolution of the issue. Are we going to reopen name suggestions? I few I had: Congo [Con for connectors] RepoMan [Repo for Repositories] ConMan? [Con for Content, but too shady] Reconto [Re for Repository, Con for Connector or Cont for Content] Reptile [Rep for Repositories, tile as an allusion to organizing or arranging things] Ralph [In honor of Ralph Kramden, bus driver on The Honeymooners. ACF is a form of software bus] Connie [Conn for Connectors] Repositor [Abbreviation of Repository, tor of connector] They may not be great, but maybe somebody can create derivative names to improve them. If there are trademark issues (e.g., Congo), maybe subtle variations (other than bolting on CF) could fix the problem. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 9:28 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Do you wish to change your vote, in that case? Karl On Tue, Sep 21, 2010 at 8:02 AM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: I'd much prefer a simple, short, name. Using a descriptive phrase as a name has these problems. Tacking on CF does indeed fix one problem, but at a high cost. That said, I am okay with a combo of a short name and a long name. So, if the short name were Ralph, the long name would be Apache Ralph Connectors Framework and we would speak of either the Apache Ralph Connectors Framework or just Ralph. Class names would begin with the capitalized short name, Ralph, and package and file names would use the lower-case, ralph as in org.apache.ralph.core.interfaces.RalphException. And upon graduation, the project would be housed at http://ralph.apache.org/. Now, I wasn't seriously considering Ralph as a name for LCF, but... it works for me. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 3:16 AM To: connectors-dev connectors-dev@incubator.apache.org Subject: Exploring ManifoldCF ramifications Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl
Re: [VOTE] Pick either Apache Connectors Framework or Apache ManifoldCF
Jack has withdrawn his vote, and I withdraw mine as well. We are going to open name nominations again, since we have apparently all had ideas since the last round that may be to people's liking. A new vote will be called to select among the new nominations, and the winner of that will go up against ACF, which is the current name as I understand it. Karl On Mon, Sep 20, 2010 at 3:46 PM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: +1 -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Monday, September 20, 2010 3:23 PM To: connectors-dev@incubator.apache.org Subject: Re: [VOTE] Pick either Apache Connectors Framework or Apache ManifoldCF Based on the level of controversy, and the necessity of settling this promptly, once and for all, I vote +1. Karl
Re: Exploring ManifoldCF ramifications
That sounds fine with me. Assuming by ACF you mean the full name Apache Connectors Framework with ACF as the local informal name, although in practice maybe everybody would refer to it, even externally, as ACF as well, as was with LCF. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 11:16 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Add: Maniplex I think we've indeed reopened discussion on this issue. I'll track these names and once again hold a vote for the purposes of selection, this time NOT including ACF. Then we will have a final vote on whether to replace ACF with the new name, whatever we come up with. Agreed? Karl On Tue, Sep 21, 2010 at 11:13 AM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: It sounds like the vote is moot pending resolution of the doc/code usage issue, but I'll go ahead and withdraw my vote at least temporarily pending further evolution of the issue. Are we going to reopen name suggestions? I few I had: Congo [Con for connectors] RepoMan [Repo for Repositories] ConMan? [Con for Content, but too shady] Reconto [Re for Repository, Con for Connector or Cont for Content] Reptile [Rep for Repositories, tile as an allusion to organizing or arranging things] Ralph [In honor of Ralph Kramden, bus driver on The Honeymooners. ACF is a form of software bus] Connie [Conn for Connectors] Repositor [Abbreviation of Repository, tor of connector] They may not be great, but maybe somebody can create derivative names to improve them. If there are trademark issues (e.g., Congo), maybe subtle variations (other than bolting on CF) could fix the problem. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 9:28 AM To: connectors-dev@incubator.apache.org Subject: Re: Exploring ManifoldCF ramifications Do you wish to change your vote, in that case? Karl On Tue, Sep 21, 2010 at 8:02 AM, Jack Krupansky jack.krupan...@lucidimagination.com wrote: I'd much prefer a simple, short, name. Using a descriptive phrase as a name has these problems. Tacking on CF does indeed fix one problem, but at a high cost. That said, I am okay with a combo of a short name and a long name. So, if the short name were Ralph, the long name would be Apache Ralph Connectors Framework and we would speak of either the Apache Ralph Connectors Framework or just Ralph. Class names would begin with the capitalized short name, Ralph, and package and file names would use the lower-case, ralph as in org.apache.ralph.core.interfaces.RalphException. And upon graduation, the project would be housed at http://ralph.apache.org/. Now, I wasn't seriously considering Ralph as a name for LCF, but... it works for me. -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 3:16 AM To: connectors-dev connectors-dev@incubator.apache.org Subject: Exploring ManifoldCF ramifications Folks, The ManifoldCF name possibility leads to some challenges as far as our documentation is concerned. I thought that it might be a good idea during the vote to explore those to see what people thought. Here are some examples of how Apache Connectors Framework might get used in text: Apache Connectors Framework is an interesting offering from Apache. ACF links repositories with search indices. That's what ACF does. The Apache Connectors Framework is a framework for repository connectors primarily. The above is not technically proper. So instead we might conceivably have done this: Apache Connectors Framework is an interesting offering from Apache. Connectors Framework links repositories with search indices. That's what CF does. The Connectors Framework is a framework for repository connectors primarily. What is the equivalent for Apache ManifoldCF? Apache ManifoldCF is an interesting offering from Apache. ManifoldCF links repositories with search indices. That's what MCF does. ManifoldCF is a framework for repository connectors primarily. Note that the difference is that we would never say, The Apache ManifoldCF... or The Apache Manifold Connectors Framework..., just ManifoldCF Would we want to use the MCF abbreviation at all? Or just convert ACF - ManifoldCF wherever it is found in documentation? Similarly, the handle acf in package and class names would need to be addressed: org.apache.acf.core.interfaces.ACFException - ? org.apache.acf.core.system.ACF - ? ...bearing in mind that you'd better choose a consistent treatment for uppercase ACF in both contexts. (FWIW, my initial thought is: org.apache.acf.core.interfaces.ACFException - org.apache.mcf.core.interfaces.ManifoldCFException org.apache.acf.core.system.ACF - org.apache.mcf.core.system.ManifoldCF) Thoughts? Karl
Soliciting more potential names for the project formerly known as LCF
We're going to have another round of name selection, based on feedback from the board. We've also pretty much decided that the last round did not have sufficient breadth, so here's an open solicitation call for potential names for the next round. So hold on to your hats, and throw your suggestions into the ring. One word of caution. Before offering a name, please do your best to make sure it meets the following criteria: (1) It's a single word (2) It's not a very commonly-used word, or a well-known place; ideally it's not a word you would find in the OED (3) It does not have negative connotations (4) A quick google with it in the context of open source shows nothing (5) No more than 4 syllables Here's what we've got so far for this round: Congo Connie Heterolink Heterosource Heteroweb Manicon Manifolio Manilink Maniplex Manisource Maniweb Multicon Ralph Reconto RepoMan Repositor Reptile Some of these don't fulfill the criteria above (e.g. Congo and Reptile), and may be stricken from the voting list as a result. But feel free to riff at will... ;-) Karl
Re: Soliciting more potential names for the project formerly known as LCF
I few more from my noon walk: Recon [Re from Repository and con from connectors] Reconn Conx [ala Connects] - treat these four as one for initial voting, then separate Conex Connx Connex Reconx [ala Repository and Connects] - ditto Reconex Reconnx Reconnex Contor [Con and tor from Connector] Contour [Contor into a word to avoid misspelling] Contango [Connectors are a bit of a dance; a term from commodities futures trading] -- Jack Krupansky -- From: Karl Wright daddy...@gmail.com Sent: Tuesday, September 21, 2010 4:27 PM To: connectors-dev connectors-dev@incubator.apache.org Subject: Soliciting more potential names for the project formerly known as LCF We're going to have another round of name selection, based on feedback from the board. We've also pretty much decided that the last round did not have sufficient breadth, so here's an open solicitation call for potential names for the next round. So hold on to your hats, and throw your suggestions into the ring. One word of caution. Before offering a name, please do your best to make sure it meets the following criteria: (1) It's a single word (2) It's not a very commonly-used word, or a well-known place; ideally it's not a word you would find in the OED (3) It does not have negative connotations (4) A quick google with it in the context of open source shows nothing (5) No more than 4 syllables Here's what we've got so far for this round: Congo Connie Heterolink Heterosource Heteroweb Manicon Manifolio Manilink Maniplex Manisource Maniweb Multicon Ralph Reconto RepoMan Repositor Reptile Some of these don't fulfill the criteria above (e.g. Congo and Reptile), and may be stricken from the voting list as a result. But feel free to riff at will... ;-) Karl
