|
2 years ago ::
May 19, 2011 - 6:36PM
#231
|
|
|
I don't want to have to copy/paste everything three or four times over, so I'm not going to do that edit right now; the noted errors have been fixed in the master copy, so they'll be corrected once I finish the last section and re-edit everything in to get the cross-referencing right.
(Emissary/Horror should already be in there--it was on the master copy already, right where it should be.)
And so people say to me, "How do I know if a word is real?" You know, anyone who's read a children's book knows that love makes things real. If you love a word, use it! That makes it real. Being in the dictionary is an artificial distinction; it doesn't make the word any more real than any other word. If you love a word, it becomes real.
--Erin McKean, Redefining the Dictionary
|
|
|
|
2 years ago ::
May 19, 2011 - 7:18PM
#232
|
- Celestial Teapots are broken!
Date Joined:
Feb 24, 2007
|
Ah, sorry about the misreport. The actual problem was that Soltari Emissary  was misspelled (it should just have one "m"). I'm able to automatically find discrepancies, but then I have to go through these by hand to see how they should be fixed, and that's where the human error can creep in.
|
|
|
|
2 years ago ::
May 20, 2011 - 12:32AM
#233
|
|
|
Ah, okay; corrected in the master now. I've also finished up black and edited that in.
And so people say to me, "How do I know if a word is real?" You know, anyone who's read a children's book knows that love makes things real. If you love a word, use it! That makes it real. Being in the dictionary is an artificial distinction; it doesn't make the word any more real than any other word. If you love a word, it becomes real.
--Erin McKean, Redefining the Dictionary
|
|
|
|
2 years ago ::
May 20, 2011 - 10:08AM
#234
|
- Celestial Teapots are broken!
Date Joined:
Feb 24, 2007
|
Okay, here are the issues I see with the black section:
(REMOVED LIST)
|
|
|
|
2 years ago ::
Jun 07, 2011 - 4:07PM
#235
|
|
|
Those errors have been corrected in the master copy; they'll be corrected on the boards copy once I finish up and re-edit everything. I've finished adding all the entries for Red and Green; just Colorless, Multicolored, and Land left.  Quick request if I could, adeyke: could you edit the big post above to either take the card tags out and/or remove the stuff I've already finished? I think the massive number of card tags it contains are eating up a ton of memory any time I try to open this page--whenever I try my browser starts slogging as though I dipped it in molasses. 
And so people say to me, "How do I know if a word is real?" You know, anyone who's read a children's book knows that love makes things real. If you love a word, use it! That makes it real. Being in the dictionary is an artificial distinction; it doesn't make the word any more real than any other word. If you love a word, it becomes real.
--Erin McKean, Redefining the Dictionary
|
|
|
|
2 years ago ::
Jun 07, 2011 - 8:41PM
#236
|
- Celestial Teapots are broken!
Date Joined:
Feb 24, 2007
|
Certainly. I removed all the monocolored entries. It seems I'd included some erroneous entries. Thank you for catching those, and sorry for the mistake. In the red/green sections, I noticed these issues: In Goblin Spelunkers  's entry, Elite Cat Warrior  is misspelled "Warriror". In Emerald Oryx  's section, Merrow Harbinger  is misnamed Merfolk Harbinger Silverback Ape  's entry is missing a close color tag after Kavu Titan. The Thicket Basilisk  < Cockatrice  comparison is missing (that had previously been present indirectly via an incorrect Infernal Medusa  comparison, so correcting that error made it not appear at all)
|
|
|
|
2 years ago ::
Jun 07, 2011 - 9:01PM
#237
|
|
|
Ah, yes, that seems to have speeded things up enormously. Thanks. No problem on the error-catches; I only just noticed Flameshot didn't hit players because I couldn't remember the card at all and went to see what it was, and after that I checked the rest of that entry and found Forked Lightning. I've corrected the mentioned issues in the master copy.
And so people say to me, "How do I know if a word is real?" You know, anyone who's read a children's book knows that love makes things real. If you love a word, use it! That makes it real. Being in the dictionary is an artificial distinction; it doesn't make the word any more real than any other word. If you love a word, it becomes real.
--Erin McKean, Redefining the Dictionary
|
|
|
|
2 years ago ::
Jun 08, 2011 - 3:27PM
#238
|
|
|
All right, all comparisons have now been added, save only Chromatic Sphere < Chromatic Star, since your opponent can't respond to the card-draw from the Sphere. There has got to be a better way of doing all those land comparisons. The sheer number of cycles and the existence of the duals absolutely buries everything in a mass of text. How about if we make entries for entire cycles and compare those cycles? Let's take Seachrome Coast / Coastal Tower / Tundra as an example--their entries would look something like this: - Coastal Tower
- Member of the Taplands Cycle. See that entry for cycle comparisons.
- Dual Land Cycle
- Consists of: Badlands / Bayou / Plateau / Savannah / Scrubland / Taiga / Tropical Island / Tundra / Underground Sea / Volcanic Island
- Seachrome Coast
- Member of the Scars of Mirrodin Land Cycle. See that entry for cycle comparisons.
- Scars of Mirrodin Land Cycle
- Consists of: Blackcleave Cliffs / Copperline Gorge / Darkslick Shores / Razorverge Thicket / Seachrome Coast
- < The Dual Land Cycle
- Taplands Cycle
- Consists of: Coastal Tower / Elfhame Palace / Salt Marsh / Shivan Oasis / Urborg Volcano
- < The Scars of Mirrodin Land Cycle (See also the Scars of Mirrodin Land Cycle )
- Tundra
- Member of the Dual Land Cycle. See that entry for cycle comparisons.
So < (and similar comparisons) in a cycle's entry would assume that each entry in the "worse" cycle is either < or <-- each entry in a better cycle. Cycles such as Ancient Amphitheater  's or Nimbus Maze  's that differ in ways other than colors of mana produced would not have cycle entries--they'd be added to individual card lists as appropriate. This would add entries, but save enormously on text and on mindspace, since there's a ton less duplication. It should also require less effort to maintain; future cycles that are comparable to existing ones won't require editing adding five, ten, or more cards in a whole cycle's worth of individual card entries--rather, an entry is made for each card and for the cycle, and then the entry for the previous cycles is added. Sanity check--does this seem like a workable system?
And so people say to me, "How do I know if a word is real?" You know, anyone who's read a children's book knows that love makes things real. If you love a word, use it! That makes it real. Being in the dictionary is an artificial distinction; it doesn't make the word any more real than any other word. If you love a word, it becomes real.
--Erin McKean, Redefining the Dictionary
|
|
|
|
2 years ago ::
Jun 08, 2011 - 6:38PM
#239
|
- Celestial Teapots are broken!
Date Joined:
Feb 24, 2007
|
To be honest, I don't like the idea. I've been making steady progress towards more automation to reduce the influence of human error. Ideally, I'd want to be able to just feed it the new Oracle dump and have it spit out the new comparisons. I've made great strides towards that, though the process still needs a lot of human input (essentially, I've moved back a step to making comparisons between abilities and then using that to automatically find comparisons between cards; I can go into more detail if you want). So the idea of introducing unofficial cycle names, treating lands differently from other cards, and applying this only to a subset of lands (normally, no distinction is made between cards that differ in mana symbols and those that differ in color words, basic land types, or some combination of the three; they're all just "color-based references") really doesn't appeal to me. I'd be much more in favor of the "See the < entries of" technology you'd suggested here. Basically, if A — B and B < C (but A isn't < C), it would just list "A <— (See the < entries of B)". That way, if we get another cycle of sometimes-ETB-tapped lands, we'd have: NewLandWU — NewLandUB, NewLandBR, NewLandRG, NewLandGW < Tundra <— (See the < entries of NewLandUB, NewLandBR, NewLandRG, NewLandGW) And Coastal Tower would just get < NewLandWU added (since the <— comparisons would already be covered by the "(See the < entries of Elfhame Palace, Salt Marsh, Shivan Oasis, Urborg Volcano)"). This wouldn't needed any special exceptions, and it would also work for, for example, the various landwalking creatures (at least, those that have — entries). The only cases where it wouldn't help would be those cases where there isn't the required — comparison, such as on the Lorwyn tribals lands or on those cases where the cycles don't overlap completely (e.g. there's no plain enemy-color ETB-tapped cycle). Also, it would mean that the <— entry of many lands would be a list of five or ten "see the < entries of" references. Let me know if that system would be acceptable. If not, we could try working on the named-cycle system to try to make that more palatable. Here are the issues I noticed in the latest additions: (DELETED LIST)
|
|
|
|
2 years ago ::
Jun 10, 2011 - 3:43PM
#240
|
|
|
To be honest, I don't like the idea. I've been making steady progress towards more automation to reduce the influence of human error. Ideally, I'd want to be able to just feed it the new Oracle dump and have it spit out the new comparisons. I've made great strides towards that, though the process still needs a lot of human input (essentially, I've moved back a step to making comparisons between abilities and then using that to automatically find comparisons between cards; I can go into more detail if you want).
So the idea of introducing unofficial cycle names, treating lands differently from other cards, and applying this only to a subset of lands (normally, no distinction is made between cards that differ in mana symbols and those that differ in color words, basic land types, or some combination of the three; they're all just "color-based references") really doesn't appeal to me.
That's definitely understandable, and I do still like the idea of using the "see < entries of".
However, as you said, even with that arrangement we'd still end up having extremely long "see" lists among the lands, and not all cycles would be fully covered. It's just not nearly as easy for a reader to follow and understand as having groups of cycles and comparing such cycles directly. I'm not going to push anything through before talking things out with you, but I do still want to find some way of working that if it's at all possible.
The things you mention as problems (unofficial cycle names, treating lands differently, and applying only to a subset) aren't set in stone; we can avoid each of them. The first we could do by not having a separate cycle entry, but putting "forms a color cycle with..." listing into the entry for the first member of the cycle, and having the cycle comparisons there. The other two could be avoided by simply not doing them--apply the same principle to the entire list where appropriate, and include other forms of color-based references. So the Signets would be such a cycle, as would the Circles of Protection, the Wards, the Laces, the neocharms, whatever. It could even possibly be used where no actual "cycle" as such was intended, but such cards have just ended up being printed over the years. (2C 2/2 landwalkers, for example)
Here are the issues I noticed in the latest additions:
All of the noted issues have been fixed, and I also got fed up with all the selecting, copying, and pasting I have to do to copy each individual section into its own post, so I've taken the radical step of putting everything into one post. The only reason it was ever split up in the first place was the old boards' character limit, and that doesn't seem to be a problem any more.
This massively simplifies the process of updating the list, since I now only need to select all/copy/paste the entire master copy into one post when making edits, instead of having to manually select each of thirteen sections and edit each into its own post. I'm not sure how your programs capture the list's contents to compare, so if this causes major problems I can split things up again, but I'd really prefer not to.
Currently there's still extra section heads from the old section divisions and a few "space reserved" markers, but those will be removed eventually; I didn't want to take them out in case you were using them as division markers in some way.
And so people say to me, "How do I know if a word is real?" You know, anyone who's read a children's book knows that love makes things real. If you love a word, use it! That makes it real. Being in the dictionary is an artificial distinction; it doesn't make the word any more real than any other word. If you love a word, it becomes real.
--Erin McKean, Redefining the Dictionary
|
|
|
~ 
~ 
~ 
~ 
