HECnet May 2013

hecnet@lists.dfupdate.se

38 participants
680 discussions

Another sillyness. More information in the nodename database on MIM.

by hvlems＠zonnet.nl

Well at least f$element comes in handy :-; Van: Brian Hechinger Verzonden: zondag 19 mei 2013 13:48 PM Aan: hecnet at Update.UU.SE Beantwoorden: hecnet at Update.UU.SE Onderwerp: Re: [HECnet] Another sillyness. More information in the nodename database on MIM. CSV is a perfectly machine readable format that can have a properly defined schema. Don't get me wrong, I like JSON and XML but I don't think they are the correct uses for this case. A few simple rules is all it takes. I'll outline a proposal later when I'm in front of a computer. -brian On May 19, 2013, at 2:20, Mark Wickens <mark at wickensonline.co.uk> wrote: > On 19/05/2013 01:25, Johnny Billquist wrote: >> After some more thinking on my part, I think I'm going to go slow on this. >> Yes, having me manage this does not scale. On the other hand, I'm not convinced it needs scaling. If updates really becomes an issue, or if people really starts asking for, and using data from this database, I look at it more as a curiosity and extension of just the basic need I had for a database for node names and node numbers, with a person attached. >> >> But I'm very interested in having more of a discussion about what could possibly be done. And I also take donations in form of code... ;-) > If the info is to be useful, other than in a TYPE INFO.TXT capacity then it needs to be in a machine readable format with a well defined schema. Two obvious choices are XML or JSON. Whilst I work more with XML in a day-to-day basis JSON is definitely more 'human-friendly'. > > I think we've got to this point in this discussion quite a few times - what is the point of taking the effort to get hecnet machine owners to provide this information? From a personal point of view I think a google map with an indication of the links between areas would be great, but there is not really any point in doing this unless we have a good initial uptake and then people keep their info up-to-date. It would be good promotional material, if that's deemed of use. > > I'm happy to define an initial schema based in the existing info file data (looking at how the data currently available could be shoehorned into a tighter schema definition) and we could go from there. > > However, I'm suspecting that there's probably half a dozen of us that would get on board with this - how many of the area operators are active on the mailing list? > > Regards, Mark. > > -- > -- > Mark Wickens > http://wickensonline.co.uk > http://declegacy.org.uk > http://retrochallenge.org > https://twitter.com/#!/@urbancamo > HECNET: http://hecnet.eu, Email: HALO::MSW >

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by wonko＠4amlunch.net

CSV is a perfectly machine readable format that can have a properly defined schema. Don't get me wrong, I like JSON and XML but I don't think they are the correct uses for this case. A few simple rules is all it takes. I'll outline a proposal later when I'm in front of a computer. -brian On May 19, 2013, at 2:20, Mark Wickens <mark at wickensonline.co.uk> wrote: On 19/05/2013 01:25, Johnny Billquist wrote: After some more thinking on my part, I think I'm going to go slow on this. Yes, having me manage this does not scale. On the other hand, I'm not convinced it needs scaling. If updates really becomes an issue, or if people really starts asking for, and using data from this database, I look at it more as a curiosity and extension of just the basic need I had for a database for node names and node numbers, with a person attached. But I'm very interested in having more of a discussion about what could possibly be done. And I also take donations in form of code... ;-) If the info is to be useful, other than in a TYPE INFO.TXT capacity then it needs to be in a machine readable format with a well defined schema. Two obvious choices are XML or JSON. Whilst I work more with XML in a day-to-day basis JSON is definitely more 'human-friendly'. I think we've got to this point in this discussion quite a few times - what is the point of taking the effort to get hecnet machine owners to provide this information? From a personal point of view I think a google map with an indication of the links between areas would be great, but there is not really any point in doing this unless we have a good initial uptake and then people keep their info up-to-date. It would be good promotional material, if that's deemed of use. I'm happy to define an initial schema based in the existing info file data (looking at how the data currently available could be shoehorned into a tighter schema definition) and we could go from there. However, I'm suspecting that there's probably half a dozen of us that would get on board with this - how many of the area operators are active on the mailing list? Regards, Mark. -- -- Mark Wickens http://wickensonline.co.uk http://declegacy.org.uk http://retrochallenge.org https://twitter.com/#!/@urbancamo HECNET: http://hecnet.eu, Email: HALO::MSW

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by mcguire＠neurotica.com

On 05/19/2013 01:30 AM, Mark Wickens wrote: I've done a lot of database work, going back to Ingres and QUEL. I'm one of those weirdos who actually enjoys databases...I think most people find database work to be dry and boring, but I find it fascinating and stimulating. I've seen DTR applications used in production but have never had any exposure at all to the software...having something to actually *do* with it, like this node database for HECnet, is great stuff, and a great way to learn. I avoided databases for a long time, but I kind of enjoy them too, although it often leaves me feeling a little dirty ;) ;) -- Dave McGuire, AK4HZ New Kensington, PA

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by mark＠wickensonline.co.uk

On 18/05/2013 16:49, Dave McGuire wrote: I've done a lot of database work, going back to Ingres and QUEL. I'm one of those weirdos who actually enjoys databases...I think most people find database work to be dry and boring, but I find it fascinating and stimulating. I've seen DTR applications used in production but have never had any exposure at all to the software...having something to actually *do* with it, like this node database for HECnet, is great stuff, and a great way to learn. -Dave Dave, I avoided databases for a long time, but I kind of enjoy them too, although it often leaves me feeling a little dirty ;) Mark. -- -- Mark Wickens http://wickensonline.co.uk http://declegacy.org.uk http://retrochallenge.org https://twitter.com/#!/@urbancamo HECNET: http://hecnet.eu, Email: HALO::MSW

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by mark＠wickensonline.co.uk

On 19/05/2013 01:25, Johnny Billquist wrote: After some more thinking on my part, I think I'm going to go slow on this. Yes, having me manage this does not scale. On the other hand, I'm not convinced it needs scaling. If updates really becomes an issue, or if people really starts asking for, and using data from this database, I look at it more as a curiosity and extension of just the basic need I had for a database for node names and node numbers, with a person attached. But I'm very interested in having more of a discussion about what could possibly be done. And I also take donations in form of code... ;-) If the info is to be useful, other than in a TYPE INFO.TXT capacity then it needs to be in a machine readable format with a well defined schema. Two obvious choices are XML or JSON. Whilst I work more with XML in a day-to-day basis JSON is definitely more 'human-friendly'. I think we've got to this point in this discussion quite a few times - what is the point of taking the effort to get hecnet machine owners to provide this information? From a personal point of view I think a google map with an indication of the links between areas would be great, but there is not really any point in doing this unless we have a good initial uptake and then people keep their info up-to-date. It would be good promotional material, if that's deemed of use. I'm happy to define an initial schema based in the existing info file data (looking at how the data currently available could be shoehorned into a tighter schema definition) and we could go from there. However, I'm suspecting that there's probably half a dozen of us that would get on board with this - how many of the area operators are active on the mailing list? Regards, Mark. -- -- Mark Wickens http://wickensonline.co.uk http://declegacy.org.uk http://retrochallenge.org https://twitter.com/#!/@urbancamo HECNET: http://hecnet.eu, Email: HALO::MSW

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by bqt＠softjar.se

After some more thinking on my part, I think I'm going to go slow on this. Yes, having me manage this does not scale. On the other hand, I'm not convinced it needs scaling. If updates really becomes an issue, or if people really starts asking for, and using data from this database, I look at it more as a curiosity and extension of just the basic need I had for a database for node names and node numbers, with a person attached. But I'm very interested in having more of a discussion about what could possibly be done. And I also take donations in form of code... ;-) Johnny On 2013-05-18 23:54, Sampsa Laine wrote: On 18 May 2013, at 23:48, Johnny Billquist <bqt at softjar.se> wrote: On 2013-05-18 23:12, Sampsa Laine wrote:

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by sampsa＠mac.com

On 18 May 2013, at 23:48, Johnny Billquist <bqt at softjar.se> wrote: On 2013-05-18 23:12, Sampsa Laine wrote:

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by bqt＠softjar.se

On 2013-05-18 23:12, Sampsa Laine wrote:

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by bqt＠softjar.se

On 2013-05-18 23:10, Sampsa Laine wrote: Always doable, just pick one box per area. Indeed. Or actually, not even one box per area is required. But it's something that needs to be agreed upon, as well as details on formats. Agreement is always a big issue. Once that is in place, and the actual data design is in place, writing code is always easy. As for easy to scrape, we had a pretty well-defined format for where the machine readable stuff starts (the content is just CSV, more or less) and ends. How much easier do you want it? :) Who knows? :-) Johnny sampsa On 18 May 2013, at 23:08, Johnny Billquist <bqt at softjar.se> wrote: On 2013-05-18 22:58, Sampsa Laine wrote: On 18 May 2013, at 18:41, Johnny Billquist <bqt at softjar.se> wrote: Or I could possible scrape files in know locations to manage the updating, if that would make more sense. (The last one would probably be really easy from my point of view...) Johnny Didn't we develop a format for this ages ago, to be stuck at the end of INFO.TXT on the public accessible dir of a machine? I don't think the current INFO.TXT files are that useful, or even easy to scrape. But that might just be me. :-) Not to mention that I have no clue which machines to scrape, if I were to do that. If I were to do it, I'd probably ask for a different format, and have a designated machine per person/area to scrape. Johnny -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol

11 years, 11 months

Another sillyness. More information in the nodename database on MIM.

by bqt＠softjar.se

On 2013-05-18 21:11, h vlems wrote: Say like each area router uses csv files :) ? csv files might work, but the area router is sometimes definitely not the right machine. Think Cisco for example... I think it both requires some more thinking, and also some more talking and experimenting. Also, how much are people really interested in this? I mean, it's a fun toy, and it could potentially be used for a lot more, but if people will not really update information, there isn't much point in implementing something. In that case just me doing the occasional updates will work better. But for it to be useful for stuff, it probably needs some kind of usable interface. You could for instance imagine combining this with a Google map, and plotting all the places where there are machines, and you could get more information about who, and what machines on the map. Cool visualization of the spread of Hecnet. Also, when people are curious about some specific thing on a certain OS, or version of an OS, they could search through the database to find who they might contact to get some help. (But that might also require the we start having email addresses in there.) Or the occasional email alert if an area looses connectivity. You can probably come up with more ideas, if you think about it some. But they are only ideas until someone actually do something. I have way too many projects already, and I am using the nodename database to create the nodename list in DECnet on MIM, as well as two web services already. I don't think I'll do much more right now. No need to push for solutions here until there at least is some request for it. Like I said, another solution would be just to have people log into MIM to update information. I can easily write a program that will allow them, and with appropriate protection. So there are many possible solutions. I'm just not sure which is the best, or even if any is meaningful to do. Johnny *Van: *Johnny Billquist *Verzonden: *zaterdag 18 mei 2013 23:08 PM *Aan: *hecnet at Update.UU.SE *Beantwoorden: *hecnet at Update.UU.SE *Cc: *Sampsa Laine *Onderwerp: *Re: [HECnet] Another sillyness. More information in the nodename database on MIM. On 2013-05-18 22:58, Sampsa Laine wrote: > > On 18 May 2013, at 18:41, Johnny Billquist <bqt at softjar.se> wrote: > >> Or I could possible scrape files in know locations to manage the updating, if that would make more sense. (The last one would probably be really easy from my point of view...) >> >> Johnny > > Didn't we develop a format for this ages ago, to be stuck at the end of INFO.TXT on the public accessible dir of a machine? I don't think the current INFO.TXT files are that useful, or even easy to scrape. But that might just be me. :-) Not to mention that I have no clue which machines to scrape, if I were to do that. If I were to do it, I'd probably ask for a different format, and have a designated machine per person/area to scrape. Johnny -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol

11 years, 11 months

← Newer
1
...
17
18
19
20
21
22
23
...
68
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

HECnet May 2013