Date: Sat, 19 Jun 93 23:39:29 EDT From: lojbab@grebyn.com (Logical Language Group) Message-Id: <9306200339.AA11594@grebyn.com> To: cowan@snark.thyrsus.com Subject: gismu list proposals, responses Cc: nsn@mullian.ee.mu.oz.au X-Mozilla-Status: 0001 I have processed most of these now, except for long lists of xrefs. Following are responses on those of your proposals that I did not essentially incorporate as you suggested. Unfortunately, I didn't start this till after I had processed a few of the changes, so I don't explicitly respond on kubli, which I resolved, but differently from what you suggested in your comments. lojbab dargu: are x2 and x3 (endpoints) necessary, given x4 as route? (also naxle, etc.) yes - 1) in practice, route may be specified either as an interval or a via 2) the from/to is a focus and may not actually be the endpoints. I95 is a road from here to NYC, but the route place would not necessarily indicate that these two points are the focus, since the road goes beyond DC and NYC. 3)like birthday with jbena, this is a commonly enough used relationship for dargu as to warrant including it. 4) cf. klama, for which the route, if expressed as an ordered set would theoretically make origina and destination unnecessary. jalra: for "cockroach" read "cockroach/termite" jalra: rationale: such are the facts; surprise! I checked, and not really. Cockroaches are an order to themselves, but are sometimes categorized as part of a super order Orthopterans whose main order is Orthoptera: grasshoppers/crickets. There are one or two other orders in the greater order, none of which are termites. However, termites are considered next most distant, decended from a roachlike ancestor. I made a change to include all of the above in jalra, but wasn't too happy about it. Lojban gismu are supposed to represent popular knowledge as well as scientific categorization. Roaches are a particular kind of pest, and the broader jalra means that we need a lujvo for that previously clear concept. I am convinced primarily because we don't have gismu for locust/grasshopper, which are probably equally important in some cultures. But this makes me want to look at the remining insects and make sure they are all covered somehow in a subcategory if they are commonly recognized enough to be talked about. Specific problem I see: beetles and true 'bugs'. Semi-related issue is a word for clam/oyster/shellfish and snail. Our broad definition of shell isn't especially useful for animal life that is often distinguished by the shell. kagni: interchange x2 and x3 places kagni: rationale: purpose more useful than charterer? Possibly, but not necessarily: the megacorporations like the Japanese have are better identified by nationality of charter than by a specific purpose, and indeed most discussion of companies in an international environment will probably have the counortant. In general, I think that tools and apparati are going to tend to have the most complex place structures, and they probably should. sakci: add "fluid x2" and renumber existing x2 and x3 sakci: for "x1" read "x1 ( in doubt. pambe: Your rewording makes it so vague, no one will know what to do with it, and could easily be covered as a trivial lujvo - pressure-cause. I agree with the commenter that said the change was going too far. I added injector as a secondary term, and put in pressure gradient in the definition, but such a gradient still requires a from and a to place, and the method by which that gradient is achieved (the x5 'tool' part of the apparatus) also seems vital. Not all pumps can pump all fiuids, so the fluid place also seems important. In general, I think that tools and apparati are going to tend to have the most complex place structures, and they probably should. sakci: add "fluid x2" and renumber existing x2 and x3 sakci: for "x1" read "x1 (agent)" sakci: rationale: sucking requires something which is sucked Yes and no. I certainly don't see x1 as clearly agentive, but there is confusion between x1 and x2. I'm rewording, but not as requested. salta: "x1 is a quantity of..." can't be mass; "set ext." is not used salta: for "x1 (mass)" read "x1"; for "x2 (set ext.)" read "x2" sanso: for "x3 (set ext.)" read "x3" Don't understand - these are not the only occurances of set. ext. vs. mass for components of a mixture, and I don't see what is distinct about them. Among the things we were trying to cover with this mass/set extent, is the possibility of an incomplete specification of ingredients. The wording should be consistent among all of them. sinxa: for "sign/symbol" read "sign/symbol/signal" changed this one - old x2 and x4 represented two different but overlapping meanings: a sign for something and a signal urging something skoto: delete "(metaphor: Gaelic/Celtic)" skoto: rationale: use le'avla for "Gaelic" (includes Eire) and "Celtic" (includes Wales, Brittany, etc.) Disagree. We want the broader meaning, if necessary at the expense of the narrower one; cf. your comment on polno. Scotland really didn't meet the criteria, but JCB had it as a prim. Note also that per my dictionary, Gaelic non-technically refers to highland Scottish unless specifically prefixed by "Irish". Now, whether extending Gaelic to Celtic is appropriate, I'm not sure. I would not have qualms with Irish = west-skoto Scottish = north-skoto Welsh = south-skoto Brittany = fraso-skoto traji: note: need to fix serai, terai, verai in cmavo list traji: reorder to x1-x4-x2-x3 to agree with zmadu and mleca No. traji is related to mutce and milxe, not to zmadu and mleca Also, per your extended comment, x4 may be a range, not necessarily a set. zgike: delete "performed by x3" zgike: rationale: music need not be performed; use cusku or tigni actually, I better accept the argument that music need not be 'composed', and removed x2. It isn't music until it is realized. I think that the performer need not be agentive (i.e. naturally occuring musical sounds), and worded it as performed/produced. xatsi, xexso, petso, femti: change "1x10**" to "1E" or "1e" Not sure why these in particular - you didn't ask to change others. Note that in some cases, a change you suggested was impossible per the LogFlash formatting of the file, and comments were added beyond column 160, which will in some way or another be merged with the rest of the text once I separate the LogFlash files from the dictionary files. --- On xrefs. Unfortunately the xrefs have proved to be popular among the community for a variety of reasons, not all of which are mutually compatible. Veijo's xrefs, for example, included some rather nebulous ties of meaning - sometimes good, but they make the xrefs act as a word-net type of thesaurus. Unfortunately the list he came up with is far too incomplete for that. Others seem to want to use the xrefs similarly, to help them find the word they want, given that they have found a related word. None of this was my original intent, which was to use contrast in the place structures to suggest what the (unwritten) differences of meaning between some rather closely related gismu are; e.g. a reference from klama to cliva might cause someone to note that the two are similar in place struccture. UNfortunatley, xrefs were added haphazrdly, and no longer are limited to thid meaning funtionc. This probably also means that the cf. abbrevaiation isn't as appropriate as q.v. or some other latinate that I don't know very well. I'm not sure that we can practically sort them out without delaying things too much. I will do what I can to the post-LogFlash files, when it will be easier, and probably go to q.v. eventually as more accurate. Eventually, I see the cross references NOT as a basis of a thesausrus, but as the basis of one of those little subnotes that I like in the Webster's New World Dictionary that I've used, that gives several related synonyms and a comparative discussion of the semantics of the terms, to indicate which is used when. That will be great to have for Lojban, but probably not till the 2nd edition will it be possible. lojbab