From william.piel@yale.edu Fri Aug 8 17:25:31 2008 Return-Path: Received: from telepathy.seas.upenn.edu (telepathy-s25.seas.private.invalid [10.0.25.97]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78LPV44014800 for ; Fri, 8 Aug 2008 17:25:31 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by telepathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78LPU85024731 for ; Fri, 8 Aug 2008 17:25:30 -0400 Received: from smtp111.sbc.mail.re2.yahoo.com (smtp111.sbc.mail.re2.yahoo.com [68.142.229.94]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78LPTt6002758 for ; Fri, 8 Aug 2008 17:25:30 -0400 Received: (qmail 34185 invoked from network); 8 Aug 2008 21:25:29 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp111.sbc.mail.re2.yahoo.com with SMTP; 8 Aug 2008 21:25:29 -0000 X-YMail-OSG: 6Y7PRVcVM1ndbNqKBV9pYwms58PtCkMI6KJYjBGtETVftscuHO5lR8DpPsieGq0JRz4OgHLm9vz0gblJbfSGUyUaUhmmKrL7kB_98iz_AQK9ILYlINq9RXK62JclFmiNuc0- X-Yahoo-Newman-Property: ymail-3 Cc: Val Tannen , Mark Dominus Message-Id: From: William Piel To: Jin Ruan In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 17:25:28 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0 X-Scanned-By: MIMEDefang 2.57 on 158.130.68.71 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ Content-Transfer-Encoding: 8bit When the user forgets his password and asks to have it email to him, can we do the following: - set the "reply to" address to "admin@treebase.org" - set a BCC to "passwords@treebase.org" - set the subject line to "TreeBASE Password" Also, can we have another function, which is "email username and password using email address". This is for the user who also forgets his username, so all he has is an email address. thanks, Bill From william.piel@yale.edu Fri Aug 8 16:46:03 2008 Return-Path: Received: from empathy.seas.upenn.edu (empathy-s25.seas.private.invalid [10.0.25.102]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78Kk3oZ008775 for ; Fri, 8 Aug 2008 16:46:03 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by empathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78Kjxfw021224 for ; Fri, 8 Aug 2008 16:45:59 -0400 Received: from smtp123.sbc.mail.re3.yahoo.com (smtp123.sbc.mail.re3.yahoo.com [66.196.96.96]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78KjxtY000680 for ; Fri, 8 Aug 2008 16:45:59 -0400 Received: (qmail 9526 invoked from network); 8 Aug 2008 20:45:59 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp123.sbc.mail.re3.yahoo.com with SMTP; 8 Aug 2008 20:45:58 -0000 X-YMail-OSG: EMLmp64VM1kMhgX3gUgn1dfPdsrDRe1IhYuMPr6zkxMIXJBZ9EWzKt_ZDtWkswTzvFbMe7pCg82igllOlqVQ1Dotx0aJiQUOP_I72hUDU4mMFfp5RUzbU1tmUrNznoEH3PPvcFOtb8O6lju6lrguTo5M X-Yahoo-Newman-Property: ymail-3 Cc: Mark Dominus , Val Tannen , Mark Dominus Message-Id: From: William Piel To: Jin Ruan In-Reply-To: <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> Content-Type: multipart/alternative; boundary=Apple-Mail-100--597011309 Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 16:45:57 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0.946 HTML_10_20,HTML_MESSAGE X-Scanned-By: MIMEDefang 2.57 on 158.130.67.158 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ --Apple-Mail-100--597011309 Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes Content-Transfer-Encoding: 8bit On Aug 8, 2008, at 2:14 PM, Jin Ruan wrote: > Version 5546 is deployed to 8ball and is ready for testing. > http://8ball.sdsc.edu:6666/treebase-web/home.html great. I'll start listing various things to fix, change, etc. See below. bp --perusing a list of studies-- • The "Study ID" column should list the legacy and new accession_numbers not the internal TreeBASE2 primary key. • Rather than have "Study Name" and "Authors", let's have Authors column, Year column, and Citation column, where the citation is the Title plus Journal plus Volume plus Pages, truncated to x number of characters so that it fits. (we should also truncate the Authors column s that they fit). The user can therefore sort by accession_number, Authors, Year, and Title/citation. If there are DOIs, clicking on the title opens another window with the doi resolver; if there is a pubmed id, clicking on the title opens another window to pubmed. If there is both, doi takes priority. • When I click the "Matrices contained in these studies" I still get a list of studies. When I click the "Trees contained in these studies" I still get a list of studies. As previously mentioned, we should have a toggle so that the user can choose to view results as a list of either studies, matrices, or trees. In other words, the basic framework is search by criteria x and view results by category y -- but where you can always go back and forth between these categories. (e.g. search for author Piel and show his trees, add new criteria that the trees must contain taxon y, but maybe now show his matrices, etc) -- Here, I've clicked the "details" button after perusing a list of studies -- • It looks to me like the authors have been migrated but that the citation is still in single-string form (i.e. the R. Page data have not been imported). Does this mean that the order of the authors is arbitrary, or does their order match the order in the citation? • Can we have some javascript to hide all but the first line of the abstract. Click on a [more] link, and the rest of the abstract is revealed • Let's have auto links to doi resolvers and pubmed pages for those papers that have doi or pubmed data • Let's have a [Download Nexus] link by each analysis step entry. This causes a nexus to be written which contains all the matrices (as data blocks) and all the trees (as tree blocks) listed with that analysis step. This file should be readable by Mesquite. • Let's have author names written out in full and make their email addresses available but not scrapable by a robot (probably impossible to guarantee, but we should show some effort to prevent email harvesting) • The analyses should list tree blocks as output, not individual trees (else if there are 10,000 trees, we don't want to force the user to view an enormous page). Have something like "Output Tree Block: 15 trees. [view trees] [list trees]" if you click view trees, a separate window opens containing PhyloWidget with a list of all the trees to view. If you click list trees, the item "expands" with a list of trees. The trees should be listed such that you have all the tree metadata: tree id, tree title, name, tree type, etc etc • If an analysis step has an list of inputs and a list of outputs, the inputs should be listed first, then the outputs • Only one bullet for "software" under analysis step, e.g.: "1. Software: PAUP, Version: 4, Commands Block: [available]" If there is no version (version = NULL) then don't show "Version:", if there are no commands, don't provide a link to view commands. • Under "Input Matrix" we should have the matrix id followed by all other metadata (data type, num char, num tax, etc) followed by links to [View Matrix] and [Download Matrix] • When I click "View all Taxa" I see a list with various numbers. Let's change the columns to be more useful. How about these columns: "Taxon Label", "Taxon", "NamebankID" and "Taxid". Anything in the last two columns has a direct link to uBio and NCBI respectively. When the mouse hovers over "Taxon Label" and "Taxon", it reveals the taxon_label_id and taxon_id respectively. This is useful to allow users to report errors to us. But we don't need to explicitly list these ids because they are otherwise not useful nor formally distributed (are formal distributed IDs are study_id, matrix_id, and tree_id) • Under the "Trees", "Matrices", and "Taxa" headings, let's use these terms: "List" means give me a list of these items, "View" means show me the items in some sort of special way -- the only "View" that we have now is probably for the trees, and clicking that should open a PhyloWidget page. • When I click "View all Trees" (which we should change to "List all Trees") the display is (1) missing metadata -- I don't see any Tree Title, etc, (2) needs reorganization so that we can fit more in the spreadsheet. I suggest that we don't have a Row_#. We should start with a checkbox or radio-button column, then have a Tree Label column (and if you hover over that the tree_id pops up), a Tree Title, then followed by the various tree metadata (tree type, etc) as well as a tree size column. All the stuff about view taxa, view tree, download nexus, etc, should be button at the bottom or top of the page, and they handle whatever has been checkmarked or radio-buttoned in the list. --Apple-Mail-100--597011309 Content-Type: text/html; charset=WINDOWS-1252 Content-Transfer-Encoding: 8bit
On Aug 8, 2008, at 2:14 PM, Jin Ruan wrote:

Version 5546 is deployed to 8ball and is ready for testing.
 http://8ball.sdsc.edu:6666/treebase-web/home.html

great. 

I'll start listing various things to fix, change, etc. See below.

bp


--perusing a list of studies--

• The "Study ID" column should list the legacy and new accession_numbers not the internal TreeBASE2 primary key.
• Rather than have "Study Name" and "Authors", let's have Authors column, Year column, and Citation column, where the citation is the Title plus Journal plus Volume plus Pages, truncated to x number of characters so that it fits. (we should also truncate the Authors column s that they fit). The user can therefore sort by accession_number, Authors, Year, and Title/citation. If there are DOIs, clicking on the title opens another window with the doi resolver; if there is a pubmed id, clicking on the title opens another window to pubmed. If there is both, doi takes priority. 
• When I click the "Matrices contained in these studies" I still get a list of studies. When I click the "Trees contained in these studies" I still get a list of studies. As previously mentioned, we should have a toggle so that the user can choose to view results as a list of either studies, matrices, or trees. In other words, the basic framework is search by criteria x and view results by category y -- but where you can always go back and forth between these categories. (e.g. search for author Piel and show his trees, add new criteria that the trees must contain taxon y, but maybe now show his matrices, etc)

-- Here, I've clicked the "details" button after perusing a list of studies --

• It looks to me like the authors have been migrated but that the citation is still in single-string form (i.e. the R. Page data have not been imported). Does this mean that the order of the authors is arbitrary, or does their order match the order in the citation?
• Can we have some javascript to hide all but the first line of the abstract. Click on a [more] link, and the rest of the abstract is revealed
• Let's have auto links to doi resolvers and pubmed pages for those papers that have doi or pubmed data
• Let's have a [Download Nexus] link by each analysis step entry. This causes a nexus to be written which contains all the matrices (as data blocks) and all the trees (as tree blocks) listed with that analysis step. This file should be readable by Mesquite. 
• Let's have author names written out in full and make their email addresses available but not scrapable by a robot (probably impossible to guarantee, but we should show some effort to prevent email harvesting)
• The analyses should list tree blocks as output, not individual trees (else if there are 10,000 trees, we don't want to force the user to view an enormous page). Have something like "Output Tree Block: 15 trees. [view trees] [list trees]"  if you click view trees, a separate window opens containing PhyloWidget with a list of all the trees to view. If you click list trees, the item "expands" with a list of trees. The trees should be listed such that you have all the tree metadata: tree id, tree title, name, tree type, etc etc
• If an analysis step has an list of inputs and a list of outputs, the inputs should be listed first, then the outputs
• Only one bullet for "software" under analysis step, e.g.: "1. Software: PAUP, Version: 4, Commands Block: [available]"  If there is no version (version = NULL) then don't show "Version:", if there are no commands, don't provide a link to view commands.
• Under "Input Matrix" we should have the matrix id followed by all other metadata (data type, num char, num tax, etc) followed by links to [View Matrix] and [Download Matrix]
• When I click "View all Taxa" I see a list with various numbers. Let's change the columns to be more useful. How about these columns: "Taxon Label", "Taxon", "NamebankID" and "Taxid". Anything in the last two columns has a direct link to uBio and NCBI respectively. When the mouse hovers over "Taxon Label" and "Taxon", it reveals the taxon_label_id and taxon_id respectively. This is useful to allow users to report errors to us. But we don't need to explicitly list these ids because they are otherwise not useful nor formally distributed (are formal distributed IDs are study_id, matrix_id, and tree_id)
• Under the "Trees", "Matrices", and "Taxa" headings, let's use these terms: "List" means give me a list of these items, "View" means show me the items in some sort of special way -- the only "View" that we have now is probably for the trees, and clicking that should open a PhyloWidget page. 
• When I click "View all Trees" (which we should change to "List all Trees") the display is (1) missing metadata -- I don't see any Tree Title, etc, (2) needs reorganization so that we can fit more in the spreadsheet. I suggest that we don't have a Row_#. We should start with a checkbox or radio-button column, then have a Tree Label column (and if you hover over that the tree_id pops up), a Tree Title, then followed by the various tree metadata (tree type, etc) as well as a tree size column. All the stuff about view taxa, view tree, download nexus, etc, should be button at the bottom or top of the page, and they handle whatever has been checkmarked or radio-buttoned in the list. 






--Apple-Mail-100--597011309-- From william.piel@yale.edu Fri Aug 8 18:53:00 2008 Return-Path: Received: from telepathy.seas.upenn.edu (telepathy-s25.seas.private.invalid [10.0.25.97]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78Mr0Li024287 for ; Fri, 8 Aug 2008 18:53:00 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by telepathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78MqxTG008935 for ; Fri, 8 Aug 2008 18:52:59 -0400 Received: from smtp107.sbc.mail.re2.yahoo.com (smtp107.sbc.mail.re2.yahoo.com [68.142.229.98]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78Mqxam008397 for ; Fri, 8 Aug 2008 18:52:59 -0400 Received: (qmail 73578 invoked from network); 8 Aug 2008 22:52:59 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp107.sbc.mail.re2.yahoo.com with SMTP; 8 Aug 2008 22:52:59 -0000 X-YMail-OSG: xEmI1NYVM1n1w4Yb772xIfkTbUVFD6fVi5uexIxBrNl7hXkpgvf5nOf2hw67Vk4iUeDUYEQm5krJse4vd9k.KCAxnhktA7sUaIT1jCQdfCAttk9y2bQ3UjWK8cf.p6SmKYw- X-Yahoo-Newman-Property: ymail-3 Cc: Val Tannen , Mark Dominus Message-Id: From: William Piel To: Jin Ruan In-Reply-To: <22483394-3AC4-48AC-A024-FF790171C0CA@yale.edu> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 18:52:58 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> <22483394-3AC4-48AC-A024-FF790171C0CA@yale.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0 X-Scanned-By: MIMEDefang 2.57 on 158.130.68.71 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ Content-Transfer-Encoding: 8bit On Aug 8, 2008, at 6:48 PM, William Piel wrote: > Can the admin person have a mechanism for fusing authors? Here's a case example: Author ids 1169 and 1360 are the same person, but in one case the middle name is separated out, the other not. I'm not sure why this would be the case because in TreeBASE 1 there is only one entry for this person. Mark: did you recreate authors in some way or where they all migrated? I'm wondering why there are two records (1169 and 1360) for one person, Jon Coddington? (or are these IDs really the author-study- association table not the author table primary key) bp From william.piel@yale.edu Fri Aug 8 16:52:15 2008 Return-Path: Received: from empathy.seas.upenn.edu (empathy-s25.seas.private.invalid [10.0.25.102]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78KqFv7009650 for ; Fri, 8 Aug 2008 16:52:15 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by empathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78KqEc7023115 for ; Fri, 8 Aug 2008 16:52:14 -0400 Received: from smtp115.sbc.mail.re3.yahoo.com (smtp115.sbc.mail.re3.yahoo.com [66.196.96.88]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78KqEUj001086 for ; Fri, 8 Aug 2008 16:52:14 -0400 Received: (qmail 20622 invoked from network); 8 Aug 2008 20:52:14 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp115.sbc.mail.re3.yahoo.com with SMTP; 8 Aug 2008 20:52:13 -0000 X-YMail-OSG: a5D7vdUVM1kjVG.EUN3hf18BkCKbGXUYVC8mSHsDHdsWGI1p7T.aMEVGM8fJhx.agfx_XAq2Z03Aji4fC4uNfyxe99oFTHaCynD.9OEvW7GMhlVki9hW2WChEut5Pmk9viI- X-Yahoo-Newman-Property: ymail-3 Cc: Jin Ruan , Val Tannen , Mark Dominus Message-Id: <15E3F09D-1B96-4B10-8ACD-F6D461A2104A@yale.edu> From: William Piel To: Mark Dominus In-Reply-To: Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 16:52:12 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0 X-Scanned-By: MIMEDefang 2.57 on 158.130.67.158 X-MIME-Autoconverted: from quoted-printable to 8bit by kinetic.seas.upenn.edu id m78KqFv7009650 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ Content-Transfer-Encoding: 8bit • Also, when listing the studies, we don't need to report the status of the studies ("Published" etc) because obviously they will all be published. • Since we have now got some migrated data to work with, we can now set the search page to only view published things. bp From william.piel@yale.edu Fri Aug 8 18:13:24 2008 Return-Path: Received: from apathy.seas.upenn.edu (apathy-s25.seas.private.invalid [10.0.25.167]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78MDODJ020063 for ; Fri, 8 Aug 2008 18:13:24 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by apathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78MDNtX009595 for ; Fri, 8 Aug 2008 18:13:23 -0400 Received: from smtp115.sbc.mail.re3.yahoo.com (smtp115.sbc.mail.re3.yahoo.com [66.196.96.88]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78MDMJI005437 for ; Fri, 8 Aug 2008 18:13:23 -0400 Received: (qmail 58579 invoked from network); 8 Aug 2008 22:13:22 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp115.sbc.mail.re3.yahoo.com with SMTP; 8 Aug 2008 22:13:22 -0000 X-YMail-OSG: EkCr_28VM1lw27prOWgBuOVeB.wMfk0IdN7uNHABBCy3aqWj8RgDY8Yj5M9is1jODMixaxi7B9wONPFyrLplIsqM9DLe02m50OPZuF91uBAlulaiBnNYPCd6.vmdgi44hOk- X-Yahoo-Newman-Property: ymail-3 Cc: Val Tannen , Mark Dominus Message-Id: From: William Piel To: Jin Ruan In-Reply-To: Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 18:13:20 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0 X-Scanned-By: MIMEDefang 2.57 on 158.130.65.101 X-MIME-Autoconverted: from quoted-printable to 8bit by kinetic.seas.upenn.edu id m78MDODJ020063 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ Content-Transfer-Encoding: 8bit In the submission management pages (when the admin person has logged in): • When the admin person clicks the "Ready State Studies" can we have two more columns: - date submission created - data submission last modified (if we are only storing one of these dates, then just show that one -- the "last modified" date is most useful though) The idea here is that the admin person will want to sort the studies to give priority attention to those that are oldest. • Can we have a button so that the admin person can delete a submission? (it would be really ideal if this delete button also produced a kind of text dump in which the basic fields/data are in the text plus encapsulated nexus, etc, and all that gets emailed to the admin@treebase.org -- thus preserving a deleted record just in case this was by mistake) • If the admin person searches for a study (e.g. under the "study management" menu, he picks "Select Studies" and then asks for the "Ready" ones) he gets a list of studies, then might click on a submission id -- suddenly everything form there on out the "Study Management" menu is missing. In other words, if the admin person wants to inspect a a submission, he suddenly is unable to do things like change submission status, etc. • The admin person has the power to edit/update User Info in the User Management section. This is great and useful. One thing though is that it looks as though the admin person cannot update anything (name, phone number, email address) without issuing a new password (because the existing password is blotted out and plus the re-type password is empty). Some possible solutions: (1) expose the actual password or (2) keep the password hidden/blotted but pre-populate the "re-type" field as well as the password field. • Can we give the admin person the power to create associate editors. i.e., in the TreeBASE User Registration page, add a pop that has two choices: Normal User and Associate Editor. The associate editor has largely the same power/role as the Admin person, except that he can't designate or revoke other associate editors. Also, it would be useful if under "Study Management > For a User" you could also search by role/ status. i.e., that lets me get a list of all users who are of type "Associate Editor". • I notice that under "Study Management > Search Submission" I can get hits for submission_ids 100, 4123, 5123 but not 41234, 51234, etc. However, when I get a hit, there doesn't seem to be a submitter or citation or anything. What am I hitting on? Are these just orphaned records? or to they exist but I can't see the data? (In other cases I can search on a submission using "Study Management > Search Submission". If I search on the id 4499, I hit on a real submission.) We should change the "submission id" searching so that it is looking for the accession_numbers (i.e. for the legacy TreeBASE1 study accession numbers and any new accession numbers for submissions since the migration) bp From william.piel@yale.edu Fri Aug 8 19:07:32 2008 Return-Path: Received: from psychopathy.seas.upenn.edu (psychopathy-s25.seas.private.invalid [10.0.25.101]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78N7W6g025847 for ; Fri, 8 Aug 2008 19:07:32 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by psychopathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78N7Ur8018870 for ; Fri, 8 Aug 2008 19:07:30 -0400 Received: from smtp102.sbc.mail.re2.yahoo.com (smtp102.sbc.mail.re2.yahoo.com [68.142.229.103]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78N7Uqk009411 for ; Fri, 8 Aug 2008 19:07:30 -0400 Received: (qmail 63171 invoked from network); 8 Aug 2008 23:07:30 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp102.sbc.mail.re2.yahoo.com with SMTP; 8 Aug 2008 23:07:30 -0000 X-YMail-OSG: _LBjsZ4VM1nkl0WT5xKTQ7UxJMolOdckW22ajbZMkpC08hMD6D5Gz7lPmxIn.cq9ZdtiC4sZPnT.YPt6zWD6HJXFriWJIwxDj56YjbMkjayIdL3xAL1YsLbGCQilQAFB3Nw- X-Yahoo-Newman-Property: ymail-3 Cc: Jin Ruan , Val Tannen Message-Id: From: William Piel To: Mark Dominus In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 19:07:29 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> <22483394-3AC4-48AC-A024-FF790171C0CA@yale.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0 X-Scanned-By: MIMEDefang 2.57 on 158.130.68.91 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ Content-Transfer-Encoding: 8bit On Aug 8, 2008, at 6:52 PM, William Piel wrote: > Mark: did you recreate authors in some way or where they all > migrated? I'm wondering why there are two records (1169 and 1360) > for one person, Jon Coddington? (or are these IDs really the author- > study-association table not the author table primary key) Ah... I think I figured this one out . In one case we have: First name = "Jonathan A" Middle name = "" Last name = "Coddington" In the other case we have: First name = "Jonathan" Middle name = "A" Last name = "Coddington" I think this happened because I had started some test submissions prior to the migration, and at that point I had created a Coddington de novo because he had not been migrated yet. It looks like all the migrated names have not had the first and middle names separated. Mark: can you parse them out such that the first whitespace encountered in the first_name is used to separate the first name form the middle name(s). Should be just a simple SQL update statement using string functions. bp From william.piel@yale.edu Fri Aug 8 20:10:45 2008 Return-Path: Received: from apathy.seas.upenn.edu (apathy-s25.seas.private.invalid [10.0.25.167]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m790AjJ1032218 for ; Fri, 8 Aug 2008 20:10:45 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by apathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m790Af8A007238 for ; Fri, 8 Aug 2008 20:10:41 -0400 Received: from smtp118.sbc.mail.re3.yahoo.com (smtp118.sbc.mail.re3.yahoo.com [66.196.96.91]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m790Afmj013745 for ; Fri, 8 Aug 2008 20:10:41 -0400 Received: (qmail 758 invoked from network); 9 Aug 2008 00:10:41 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp118.sbc.mail.re3.yahoo.com with SMTP; 9 Aug 2008 00:10:40 -0000 X-YMail-OSG: 3LHi5ucVM1kjAZdWoHDRL3Pkg6NAkIR9TyVhI_QRKP_XM3Q2t.nTXER3zQ_teTZucdYjIfQMCuL8X8CMxY0uhyAmGp0OsRb9jTGSga0eSHBndNzj7JJFSvuD_C8cHcFYHbXz3r.uEuPWDoXsOf6ChAQW X-Yahoo-Newman-Property: ymail-3 Cc: Val Tannen , Mark Dominus Message-Id: <9138E251-224D-4B7E-8533-87C6493D1810@yale.edu> From: William Piel To: Jin Ruan In-Reply-To: Content-Type: multipart/alternative; boundary=Apple-Mail-159--584728835 Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 20:10:39 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> <22483394-3AC4-48AC-A024-FF790171C0CA@yale.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0.63 HTML_MESSAGE,HTML_SHORT_LENGTH X-Scanned-By: MIMEDefang 2.57 on 158.130.65.101 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ --Apple-Mail-159--584728835 Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes Content-Transfer-Encoding: 8bit Some of these comments have probably been reported before... but I'll repeat them just in case: • When TreeBASE builds together / assembles a citation (e.g. in a submission summary) can we make it so that if the issue number is missing, the function does not print the parentheses -- ie. "()"? • Attached is a tree file downloaded from a submission I'm testing (Submission: 7942). It has the following syntax errors: - the command "LINK" needs to be terminated with a semicolon ie change the line to this: "LINK TAXA = Untitled_Taxa_Block;" - likewise the TREE command needs to be terminated with a semicolon - the tree name ("Fig. 311") needs to be encapsulated following the nexus token rules. i.e.: for when there are legal characters only: "TREE Fig._311 = [&R] ((Cra..." for when there is at least one illegal character: "TREE 'Fig. 311-2' = [&R] ((Cra..." • When I downloaded a nexus file for a data matrix (in this case M3321), there were the following problem: - if "BEGIN CHARACTERS;" is used, then there must be at least one "BEGIN TAXA;" block. Do it just the way you've done with the tree file download -- i.e. give it a TITLE to the TAXA block and give the CHARACTER block a LINK TAXA command. • When I did a "View Rows" for matrix 3321 of submission 7942, nothing happened for about a minute, and then I got this message: Safari could not open the page “http://8ball.sdsc.edu:6666/treebase-web/user/matrixRowList.html?id=3321 ” because the server is not responding. • TaxonFinder is failing to find all but two of the names in submission 7942. That's obviously not our fault -- it's uBio's flaky web service, because all of the names in this submission are clearly in uBIO. But it does underscore the need for manual ways of entering namebankids by the submitting author. It also underscores the need for the admin person (or associate editors) to have the ability to "rewire" some of the taxon_label - taxon_variant - taxa relationships. For example, in the case of "Archaea", there is only one namebankid: 229973, but it is both a bacterial superkingdom and a spider genus. So we need the ability to create two taxon_variants that map to two different taxon records, one for the bacteria and one for the spider. This is not something that our taxonomic intelligence is smart enough to do by itself -- this has to be done by hand by the admin or editors using a non-frills but functional interface. • By the way, what happens when a taxon_label matches two or more existing taxon_variant records? i.e., let's say that I have two separate records for "Archaea" -- one for the spider genus and one for the bacteria superkingdom. What happens when a new submission is made with a taxon_label called "Archaea". Remember, in my mockup (http://www.waterflea.org/tbn/ username: cgi, password: sdsc) I give the user a choice (using radio buttons) between any of the taxon variants returned from uBIO -- and they always have the choice to pick "no match". In the "Validate Taxon Labels" page, can we also have popup lists for each match, so that in instances where there are more than one match, the user can choose between them; and in instances where there is a false match, the user can always say "no match" (and then the taxon_label records is not joined with a taxon_variant record)? regards, Bill --Apple-Mail-159--584728835 Content-Type: multipart/mixed; boundary=Apple-Mail-160--584728833 --Apple-Mail-160--584728833 Content-Type: text/html; charset=WINDOWS-1252 Content-Transfer-Encoding: 8bit
Some of these comments have probably been reported before... but I'll repeat them just in case:

• When TreeBASE builds together / assembles a citation (e.g. in a submission summary) can we make it so that if the issue number is missing, the function does not print the parentheses -- ie. "()"? 

• Attached is a tree file downloaded from a submission I'm testing (Submission: 7942). It has the following syntax errors:

- the command "LINK" needs to be terminated with a semicolon

ie change the line to this: "LINK TAXA = Untitled_Taxa_Block;"

- likewise the TREE command needs to be terminated with a semicolon

- the tree name ("Fig. 311") needs to be encapsulated following the nexus token rules. 

i.e.:

for when there are legal characters only: "TREE Fig._311 = [&R] ((Cra..."
for when there is at least one illegal character: "TREE 'Fig. 311-2' = [&R] ((Cra..."

• When I downloaded a nexus file for a data matrix (in this case M3321), there were the following problem:

- if "BEGIN CHARACTERS;" is used, then there must be at least one "BEGIN TAXA;" block. Do it just the way you've done with the tree file download -- i.e. give it a TITLE to the TAXA block and give the CHARACTER block a LINK TAXA command. 

• When I did a "View Rows" for matrix 3321 of submission 7942, nothing happened for about a minute, and then I got this message:

Safari could not open the page “http://8ball.sdsc.edu:6666/treebase-web/user/matrixRowList.html?id=3321” because the server is not responding.

• TaxonFinder is failing to find all but two of the names in submission 7942. That's obviously not our fault -- it's uBio's flaky web service, because all of the names in this submission are clearly in uBIO. But it does underscore the need for manual ways of entering namebankids by the submitting author. It also underscores the need for the admin person (or associate editors) to have the ability to "rewire" some of the taxon_label - taxon_variant - taxa relationships. 

For example, in the case of "Archaea", there is only one namebankid: 229973, but it is both a bacterial superkingdom and a spider genus. So we need the ability to create two taxon_variants that map to two different taxon records, one for the bacteria and one for the spider. This is not something that our taxonomic intelligence is smart enough to do by itself -- this has to be done by hand by the admin or editors using a non-frills but functional interface.

• By the way, what happens when a taxon_label matches two or more existing taxon_variant records? i.e., let's say that I have two separate records for "Archaea" -- one for the spider genus and one for the bacteria superkingdom. What happens when a new submission is made with a taxon_label called "Archaea". 

Remember, in my mockup (http://www.waterflea.org/tbn/ username: cgi, password: sdsc) I give the user a choice (using radio buttons) between any of the taxon variants returned from uBIO -- and they always have the choice to pick "no match". 

In the "Validate Taxon Labels" page, can we also have popup lists for each match, so that in instances where there are more than one match, the user can choose between them; and in instances where there is a false match, the user can always say "no match" (and then the taxon_label records is not joined with a taxon_variant record)?

regards,

Bill




--Apple-Mail-160--584728833 Content-Disposition: attachment; filename=T4801.zip Content-Type: application/zip; x-unix-mode=0644; name="T4801.zip" Content-Transfer-Encoding: base64 UEsDBBQACAAIAFybCDkAAAAAAAAAAAAAAAAJABAAVDQ4MDEudHJlVVgMAFbWnEhA1pxI9QH1AXVW bW/iOBD+nl8xp5NWIHkjaLu7fdHqRIG2XEuKGqrrqVpVJpkSi8SObKdt/v1NgCTE28sXyHhen/Hz wJ/B9Okx9LznZSIMxNxyMGjhndOLepep4jHG8KpVBkuNeDkKpww4aEy5FUrydBuz4gZBvUKelKla o0QrIthI9Z5ivEa/CYWE8q4QJZgiz5W2lHtVgk0QgvCKwQ3Xb1zH8CjFG2ojbMngX55ixxBOwjFw SV5jmPA3YXxYpFh1ECuQylJzmXpDykoT8ahuI0Npd4Nsy+FHYeBVpOh7k3ZQJWFUrAtjYXDK4Ggw OL2Ao+Pzo+9wPV96zRh/wfDs7ORrde55oS3ikoq+okYZ4bm3IGykiDYQ+AzGKo6FXFvK/Lc/IsOV MhY1PPgP9PIFrrUw7yqNYexP/Srt0IcwF1JSNgtzpfNEEaalV01cdb7YY1xWgN/wvDqUWMXEhJAP owy1iLiEeWGwyCBQb8Jyi4bB8WD4vdc/94Zffxz7deOzCfyEH2cnR94vz7ucXs8CWI6eRhce7J/l bHk3BVqBFZaAfFnyD/5ymapo07hMZvNpEM7uAwgo9ufJcRs8erobXU7vwsZSPTdlrqJEpIXpmKcR NUrwCPrsHFxrHhdyXaRd80JwqbqmG8I941J0rSPaqP6s4jLhIu76XolUGOqja70tNkW6+S1xGJVW xdhNGhL8WjiVJrrMCtONvlMfykSYOvETsS6IQU4lXCMNIZRwy9Hlj6zGQpdd+61YaZ45+CzoNkVO a2Ouuci4dUYOVN71u4+SklBE/XlnXeukNLHrOecp9UrXeeWCs3UW0tlanhP3c6IOOq09GkXUdbx1 lHB0CmJUmowXUYLSXciSWIJp6ha9KSwVdIa5J62ziYvHP1zYjTvjEq3ma8lt0rWPNTeGS54LB1QC ZOX2Flpcq7jMExcnFFK5GSZEl9KZYkxI/552pLlExzYXGVrHtqAL8Qma3jSYXDQa8TCdhrVI7CRi TtR9oVgjMkJQFealEkxTO93NgtuttJDe/L+YVHmJg2sfjodD8nz+8vALer0D+FhvPwfr9WpAWL2L fmXcAcIaECrjfk76Vm+YNVul8/Y2sF47flVq98W5Sf3qoVz19tjBxip7r1ER1ipHdVCRivVaxpGp 4Q/bc4b8GkKwQ87si+45vJ10pxSsow7kc6AJrNaBfccNiWm4Q4KxPam2XrWEsVaiqq52GsZq1evX MLR6zg41vH9+5A/ouJVvtpNsimmEmto4UGe2U+R++5wP/G/sgFXVe997/gPafxU0KcIspttycjoY Av2O7a7qf1BLBwjGe4ac0gMAAN0IAABQSwMECgAAAAAAaZsIOQAAAAAAAAAAAAAAAAkAEABfX01B Q09TWC9VWAwAVtacSFbWnEj1AfUBUEsDBBQACAAIAFybCDkAAAAAAAAAAAAAAAAUABAAX19NQUNP U1gvLl9UNDgwMS50cmVVWAwAVtacSEDWnEj1AfUBY2AVY2dgYmDwTUxW8A9WiFCAApAYAycQGwFx KRCD+MsZiAKOISFBDMFW86A65gAxN5oSRoS4eHJ+rl5iQUFOql5IakWJa15yfkpmXjpQ0tfR2dMv xD/Yw9oAAFBLBwjPUiywWwAAAKcAAABQSwECFQMUAAgACABcmwg5xnuGnNIDAADdCAAACQAMAAAA AAAAAABApIEAAAAAVDQ4MDEudHJlVVgIAFbWnEhA1pxIUEsBAhUDCgAAAAAAaZsIOQAAAAAAAAAA AAAAAAkADAAAAAAAAAAAQP1BGQQAAF9fTUFDT1NYL1VYCABW1pxIVtacSFBLAQIVAxQACAAIAFyb CDnPUiywWwAAAKcAAAAUAAwAAAAAAAAAAECkgVAEAABfX01BQ09TWC8uX1Q0ODAxLnRyZVVYCABW 1pxIQNacSFBLBQYAAAAAAwADANQAAAD9BAAAAAA= --Apple-Mail-160--584728833 Content-Type: text/html; charset=US-ASCII Content-Transfer-Encoding: 8bit
--Apple-Mail-160--584728833-- --Apple-Mail-159--584728835-- From william.piel@yale.edu Fri Aug 8 18:49:02 2008 Return-Path: Received: from psychopathy.seas.upenn.edu (psychopathy-s25.seas.private.invalid [10.0.25.101]) by kinetic.seas.upenn.edu (8.13.6/8.12.8) with ESMTP id m78Mn29n023903 for ; Fri, 8 Aug 2008 18:49:02 -0400 Received: from snowball.pcbi.upenn.edu (SNOWBALL.PCBI.upenn.edu [128.91.62.2]) by psychopathy.seas.upenn.edu (8.13.6/8.13.6) with ESMTP id m78Mn1Zi015141 for ; Fri, 8 Aug 2008 18:49:01 -0400 Received: from smtp119.sbc.mail.re3.yahoo.com (smtp119.sbc.mail.re3.yahoo.com [66.196.96.92]) by snowball.pcbi.upenn.edu (8.12.11.20060308/8.12.11) with SMTP id m78Mn1QX007980 for ; Fri, 8 Aug 2008 18:49:01 -0400 Received: (qmail 39427 invoked from network); 8 Aug 2008 22:49:01 -0000 Received: from unknown (HELO ?10.0.1.196?) (wpiel@snet.net@76.250.193.172 with plain) by smtp119.sbc.mail.re3.yahoo.com with SMTP; 8 Aug 2008 22:49:00 -0000 X-YMail-OSG: DPFA.iQVM1lEpxTSF.zdo_EY0GWH_Fv3LXHmqtIRCiQccKdkS9hcdRjhNN13mOmWo_a0YnnT6A.YWXPftD24Il0PANU3Bzgm9XoKZ6Q1eA-- X-Yahoo-Newman-Property: ymail-3 Cc: Val Tannen , Mark Dominus Message-Id: <22483394-3AC4-48AC-A024-FF790171C0CA@yale.edu> From: William Piel To: Jin Ruan In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Mime-Version: 1.0 (Apple Message framework v926) Subject: Re: Please continue to test searching Date: Fri, 8 Aug 2008 18:48:59 -0400 References: Your message of "Thu, 07 Aug 2008 23:39:17 PDT." <845AADAC1106E44996327D62097E4C6B010D32B2@et.ad.sdsc.edu> <20080808070958.22845.qmail@plover.com> <845AADAC1106E44996327D62097E4C6B010D32F2@et.ad.sdsc.edu> X-Mailer: Apple Mail (2.926) X-Spam-Status: 0 X-Scanned-By: MIMEDefang 2.57 on 158.130.68.91 X-Evolution-Source: imap://mjd@mjd.mail.seas.upenn.edu/ Content-Transfer-Encoding: 8bit Can the admin person have a mechanism for fusing authors? Currently, this is a mechanism for fusing users, which is great, but there will also be instances where one author has two or more author records. I suppose a fairly complicated way is for the admin person to go to the studies that use one instance of an author, delete this author from the citation, reenter the last name but this time pick the other instance of this author, etc. (if an author entry is finally orphaned as a result of this process, does it say there or does the last instance of it get deleted? Ideally the last instance should be deleted) A simpler way to fuse authors would be useful, but not a really high priority item. bp