ThePortForum.com

jdaw1 wrote:

[url=http://www.theportforum.com/viewtopic.php?p=90612#p90612]Here[/url] PhilW wrote:

jdaw1 wrote:The placemat code allows the adding of an annotation to glasses (GlassesAnnotations). After a blind tasting, should the placemat then be ‘changed’ to annotate what was what?

No. Final version of placemats is final version, including any errors. Scoresheet (including answers) is Scoresheet. Post-event update of placemats bad.

Having thought about this more, I disagree with Phil.

I’m about to make the decanter labels for the tasting of Sweet-Spot Vintages. When they’re made, I’ll change the placemats to /DecanterLabelsNumCopies 0 def. When Mike prints on the day, with whoever is and isn’t coming appropriately altered, he won’t be wasting the decanter-label pages. After the tasting /DecanterLabelsNumCopies 1 def will be reverted, and that will be the ‘final’ version.

This seems reasonable, at least to me. But it firmly clashes with Phil’s purism.

Further comment?

I would create one version of the placemats, with decanter-labels set to be included.
When printing, I would print the pages I want. Job done, no faffing, one version.

PhilW wrote:I would create one version of the placemats, with decanter-labels set to be included.
When printing, I would print the pages I want. Job done, no faffing, one version.

But what if responsibility for printing is likely to be split amongst people? I printing a few (decanter labels) on my home printer; Mike doing the bulk. Would the remaking of the file be easier for Mike, or for whomsoever is sub-delegated the task?

Are the computer-competent people having a trust issue? When writing robust software, sub-routines are not trusted to “alter an array, but not in any way that will hurt”. Instead the whole thing would be const, and then the sub-routine just can’t damage it at all. Much more robust.

I’m saying that I want to be able to change, but not in a way that would affect printing. Are the programmers not confident of the robustness?

jdaw1 wrote:Are the computer-competent people having a trust issue?

jdaw1 wrote:Are the programmers not confident of the robustness?

Yes it is a trust issue, though of the users/usage rather than the code. That said, it does depend on the purpose of archiving the placemats electronically.

From a software-centric perspective, best practise is for every software executable version released to have the source from which it has been created marked/identifiable within revision control, ideally along with meta-data identifying all tools and their versions used to create the executable from the source. Allowing release of the binary, and then release of the source "tweaked slightly but in a way that shouldn't affect anything" is a total no-no (since for any debug or modification your starting point is not what you think it is).

In placemat land, the placemats produced (by printing) have a file from which the printing was performed, which have a source .ps from which they were created. While it is possible for the meta-data to change (I only printed pages 10-15, or I scaled them etc), if the source changes, then both sources should be saved. In the specific example above, the .ps source is being changed; your argument is that the simple source/parameter change you are making between your two versions of source/pdf probably doesn't/shouldn't affect the page content, only which pages are produced; I don't like "probably" and prefer rigour.

If the goal of placemat archive is to provide a guide/impression, rather than the precise placemats and source used, then rigour could be compromised, though I am not persuaded to do so.

PhilW wrote:If the goal of placemat archive is to provide a guide/impression, rather than the precise placemats and source used, then rigour could be compromised, though I am not persuaded to do so.

The paper should be precisely correct, as it is the paper that is at the tasting. The source (of paper which is PDF, or of PDF which is PostScript) does not appear at the tasting.

jdaw1 wrote:

PhilW wrote:If the goal of placemat archive is to provide a guide/impression, rather than the precise placemats and source used, then rigour could be compromised, though I am not persuaded to do so.

The paper should be precisely correct, as it is the paper that is at the tasting. The source (of paper which is PDF, or of PDF which is PostScript) does not appear at the tasting.

Agreed; however the link to the placemats (in pdf format) is maintained, and should (imo) be the source used to generate (print) the paper placemats used at the tasting. Your suggestion is that the linked pdf could be based on postscript generated after the event; it would then be representative, but not guaranteed to generate identical results (in the sense that any change in source has the potential the change the output, though it may not if well-managed).

PhilW wrote:Agreed; however the link to the placemats (in pdf format) is maintained, and should (imo) be the source used to generate (print) the paper placemats used at the tasting. Your suggestion is that the linked pdf could be based on postscript generated after the event; it would then be representative, but not guaranteed to generate identical results (in the sense that any change in source has the potential the change the output, though it may not if well-managed).

We are edging closer to agreement.

What if my policy were that, post-pasting, the only parameters that may be changed are those that cannot affect the printout (except the log page)? These are the following (list might be extended later — but subject to same cannot-affect criterion).
• ExternalLinks (perhaps to add a link to the review thread);
• GlassesAnnotations (errata and corrigenda, perhaps re mis-descriptions);
• LogThisExtra (adds text to the log page — sundry message can go here);
• LicensingAgreementTextPlacemats (only for more generous licensing); and
• LicensingAgreementLinkPlacemats (ditto).

By design of the code, none of these can have any affect on the printout of any page except the log page (which I’m deeming not to be part of the placemats for these purposes).

Does that policy-and-code provide enough reassurance?

jdaw1 wrote:What if my policy were that, post-pasting, the only parameters that may be changed are those that cannot affect the printout (except the log page)? ... Does that policy-and-code provide enough reassurance?

From a formal perspective it's still definitely wrong (no change to .ps/.pdf for archived copy should be allowed once printout is produced from .pdf produced from .ps) and I would much favour the alternative of "just print the pages you want" rather than "have two different source files and hope/intend there is no error such that they would not both produce the same visual output". However, It would represent a pragmatic nominally least-risk method to allow post-tasting modification of .ps/.pdf with least change of unintentional visual change consequence, if one is prepared to breach the primary principle that the archived copy should be precisely what was used to generate the paper copy.

Glenn? Are you as persuaded as Phil?

jdaw1 wrote:Glenn? Are you as persuaded as Phil?

Given your phrasing, I should re-iterate that I am still not persuaded that we should breach the primary principle that the archived copy should be precisely what was used to generate the paper copy (I think we should not). However were you to decide to do so, the approach is subsequently sensible.

PhilW wrote:

jdaw1 wrote:Glenn? Are you as persuaded as Phil?

Given your phrasing, I should re-iterate that I am still not persuaded that we should breach the primary principle that the archived copy should be precisely what was used to generate the paper copy (I think we should not). However were you to decide to do so, the approach is subsequently sensible.

My phrasing was consistent with you being not more than slightly persuaded.

PhilW wrote:

jdaw1 wrote:Glenn? Are you as persuaded as Phil?

Given your phrasing, I should re-iterate that I am still not persuaded that we should breach the primary principle that the archived copy should be precisely what was used to generate the paper copy (I think we should not). However were you to decide to do so, the approach is subsequently sensible.

Given this answer, yes. I'm a as (not) persuaded as Phil.

But as previously stated, I'm a theoretical participant since I do not use the archived files. So if those who use them wish to alter the files prior to preservation, more power to them.

Perhaps a more formal source code management system is in order?

Glenn E. wrote:Perhaps a more formal source code management system is in order?

There are two levels of “source code”.

1. You might mean the program itself, as I and only I maintain. This question was asked, and you concurred with the negative.

   [url=http://www.theportforum.com/viewtopic.php?t=175&p=80804#p80804]Here[/url], on Monday 11th August 2014, Glenn E. wrote:

   PhilW wrote:

   jdaw1 wrote:Do any programmers know whether there would be sufficient advantages in moving my code to SourceForge.net? My prior is ‘no’, but I’m willing to be persuaded otherwise.

   I can't see a significant advantage. The primary benefits of using Sourceforge are to be able to share your code in an online public repository, facilitating multi-user development and version control. Given that your code is a single file, you already have a web server where you make the file public, and you probably want to maintain control of changes, this would seem to offer minimal benefit at this time.

   +1

2. So perhaps you mean the particular parameters for each set of placemats. You appear to be proposing using an administrative system to keep every draft .ps/.pdf. But, re the ‘final’ PDF, you said that you “do not use the archived files”. So why save every draft? I think far too much hassle for about zero gain.

Summary: no.

Post-Event Placemat Changes: The Policy

Before a tasting placemats change: people are in then out, bottles are promised and the substituted. There can be many drafts.

Then the paper print of the placemats is used at the event. So after the event there should be no change in that which would print from the PDF (excepting the log page, deemed not to be part of the placemats for these purposes). But there may be changes in the digital information that doesn’t print. To ensure that there aren’t accidental changes in that-which-would-print, after the event the only parameters that may change are those that don’t affect the printable parts of the PDF.

These parameters are the following (with parenthetical example reasons for a change).
• ExternalLinks (perhaps to add a link to the review thread, or to a website relevant to a revealed theme).
• GlassesAnnotations (corrigenda and errata, perhaps re mis-descriptions, and de-blinding).
• LogThisExtra (to add a message to the log page).
• LicensingAgreementTextPlacemats (only for more generous licensing).
• LicensingAgreementLinkPlacemats (ditto).

Discussion summary: this not the purist policy, because it relies on the operator not changing things that should not be changed. But within the limits of the formality and informality of the placemat making process, it is a practical compromise, and a practical compromise that is unlikely to do any harm.

jdaw1 wrote:Summary: no.

I believe that is the correct conclusion. I was but suggesting that it might be appropriate to consider the question again.

Considered and rejected.

There is a page type we almost never use: Decanting Notes.


Is this because it is useless and should be removed from the code? Or because everybody had forgotten, and, now reminded, will be clamouring for it? Or something else?

The most recent placemats with DN pages were AHB’s 1963 Quinquagenary in October 2013 (pdf), the Taylor vertical in March 2013 (pdf), and several the previous year including the Graham vertical in November 2012 (pdf). The only record of their use is at the matrix tasting in October 2011 (thread).

(The question is asked because a small improvement was recently made to the vote-recorder pages — small, trust me — which should be echoed to the decanting-note pages. Which raised the question of whether the DN pages have any purpose.)

I think we should keep these. They are useful for very large tastings, particularly when we remember to print them.

We never use them, but it's easy to not print them so I don't really care whether or not they remain in the code.

Basically What Derek and Glenn said. I rarely use them but when I need to its nice to have them.


Sent from my iPhone using Tapatalk

I'd like to keep them in the deck. We did use them at the Quinquagenary to make a record of the bottler and time of decanting. One or two more detailed notes were also taken.

DRT wrote:I think we should keep these. They are useful for very large tastings, …

AHB wrote:I'd like to keep them in the deck.

Consensus: they will be kept in the deck.

DRT wrote:… particularly when we remember to print them.

Note to self: remember. Or automate. Would it be sensible for them appear by default iff there are ≥15 wines (/DecantingNotesNumCopies Circlearrays length 15 ge {1} {0} ifelse def)?

Edit: the default is now /DecantingNotesNumCopies Circlearrays length 15 ge {1} {0} ifelse def.

I've never used them; Cork details is probably not really needed any longer, since we have the cork display. The only part which seems potentially useful is to have a sheet for recording decant times when decanting centrally (I have made notes of times on a scrap of paper when decanting like this previously). For anything other than vertical, unlikely to be useful since people will all decant at home and usually are asked decant time at the tasting which would be noted on normal sheets (possibly an argument for a handy space for it on the normal tasting notes, so we can use/ignore it as per the eye/nose/taste sections).

PhilW wrote:(possibly an argument for a handy space for it on the normal tasting notes, so we can use/ignore it as per the eye/nose/taste sections).

You have been ignoring it: since mid-February 2008 the code has contained /TastingNotesColumnHeadings [ (Times) (Eye) (Nose) (Mouth) (Score) ] def.

I have sympathy with your opinion that DNs aren’t needed, but consensus was the other way.

The 2015 Royal Academy Summer Exhibition:


And an extract from the placemats for the Flight of 1966 on 22^nd April 2014.