|
|
| Author |
Message |
macs none

Joined: 26 Feb 2011 Posts: 30 Location: UK
|
Posted: Jul 13, 2011 2:41pm Post subject: Help me find out how many nets use atheme and anope |
|
|
Hey There i'd like to know how many of the nets on here use Atheme and how many nets use anope
-Tom |
|
| Back to top |
|
 |
katsklaw Guru

Joined: 28 Jun 2004 Posts: 1614 Location: Somewhere you're not.
|
Posted: Jul 13, 2011 9:25pm Post subject: |
|
|
| You can't .. SearchIRC has been specifically designed so users can't search by software package. It's felt that it would prove a rather useful tool in the event a malicious user wanted to attack/exploit all nets that use X software. |
|
| Back to top |
|
 |
chaz Idler

Joined: 15 Jun 2005 Posts: 304
|
Posted: Jul 15, 2011 12:59am Post subject: |
|
|
I can't speak for nets on here, or for unique (ie per network) downloads but I do have download stats for anope if you are interested...
Averages about 100 downloads a day (3k a month) of file releases, not sure on the repositories. |
|
| Back to top |
|
 |
PingBad Post Whore

Joined: 05 Feb 2005 Posts: 3027 Location: New Zealand
|
Posted: Jul 15, 2011 2:24am Post subject: |
|
|
| while katsklaw is right, if any network feels the need to say that they're an atheme user or anope user then they may do so at their own discretion (version of either need not be mentioned) - outside of that, I can't see macs getting any viable results of their survey |
|
| Back to top |
|
 |
katsklaw Guru

Joined: 28 Jun 2004 Posts: 1614 Location: Somewhere you're not.
|
Posted: Jul 15, 2011 1:54pm Post subject: |
|
|
| chaz wrote: | I can't speak for nets on here, or for unique (ie per network) downloads but I do have download stats for anope if you are interested...
Averages about 100 downloads a day (3k a month) of file releases, not sure on the repositories. |
I really don't think that having the download statistics is going to be helpful anyway.
ABC having 10,000 downloads and XYZ having 5,000 downloads doesn't indicate that ABC is used on more nets. 4,000 of XYZ's 5,000 could be from nets that tried ABC and didn't like it and vice versa.
There is no telling how many nets are still on old versions and don't plan to upgrade. At the same time some could down load one or both to compare then eventually choose something completely different, like X3 or GNUworld. Some people could have downloaded the same version multiple times for various reasons.
Example: Anope 1.7.21 had something like 15,000 downloads although It's doubtful there is anything close to 15,000 nets let alone using Anope. |
|
| Back to top |
|
 |
chaz Idler

Joined: 15 Jun 2005 Posts: 304
|
Posted: Jul 16, 2011 5:52am Post subject: |
|
|
I didn't say they were going to be helpful, I said if interested.
I agree with what you're saying and I don't believe I said anything to the contrary or to imply anything other than providing the information. |
|
| Back to top |
|
 |
Trixar_za Eleet

Joined: 10 Dec 2006 Posts: 624 Location: South Africa
|
Posted: Jul 16, 2011 2:23pm Post subject: |
|
|
Let's try to be logical now. We can't gain the stats of which server uses which services package... but we can find out the most common IRCds:
http://searchirc.com/ircd-versions
It tells us the most common used IRCds are UnrealIRCd, InspIRCd, ratbox-ircd forks and u2 forks.
So let's break that down using the IRC Wiki's own network listing (because we allow people to list their services package). Out of the 44 listed networks, 24 listed themselves as Anope users, 12 listed themselves as Atheme users and the remaining 18 either didn't list one or just is using something else. Let's ignore that 18 and focus on Atheme vs Anope ration. Going on my stats, for every 2 Anope users there is is 1 Atheme user. So it's a Ratio of 2:1.
That's a very rough estimate, but I think that should be enough to get you going. You should also assume that more servers running UnrealIRCd would be more likely to use Anope, where as people using either InspIRCd or ratbox-ircd forks (like charybdis, shadowIRCd, ircd-seven - basically TS6 servers) are more likely to use Atheme than not. u2 forks tend to use a variation of X or X3 as their preferred Services package.
Either way, Anope still will be in the majority - even if it's closely followed by Atheme  |
|
| Back to top |
|
 |
Trixar_za Eleet

Joined: 10 Dec 2006 Posts: 624 Location: South Africa
|
Posted: Jul 16, 2011 3:05pm Post subject: |
|
|
Ok, doing actual maths provided by my stats:
1010 = Atheme
3029 = Anope
2019 = Other Services Package
-------------------------------------
6058 = How many Actively listed networks on SearchIRC
SearchIRC provided the base number and IRC-Wiki's rough estimates provided the break up of Services packages as 3/6 for Anope, 2/6 for Other and 1/6 for Atheme. This is probably not real world numbers, but a very rough estimated based on a very small test group, but it should provide you with the rough numbers your looking for... |
|
| Back to top |
|
 |
katsklaw Guru

Joined: 28 Jun 2004 Posts: 1614 Location: Somewhere you're not.
|
Posted: Jul 16, 2011 8:27pm Post subject: |
|
|
Trix, That doesn't really provide any factual figures at all. It's simply using your extremely small base and assuming the ratio is a constant which it isn't since many networks existed before Atheme became popular. So that isn't logical at all. Also, there are not 6000+ ACTIVE networks, that is a total number of networks in the data base both inactive and active. 4 of which are mine, 2 used Anope, 1 used GNUWorld and 1 used X3. I did have a testnet running Atheme for about a week but it was never indexed.
You can't use small numbers to accurately figure averages, the more data you have the more accurate the mean number. Case in point if I drive 2 miles down hill right after I reset my trip information, my truck will say it gets 30mpg. However, I currently have 175,000 miles on the trip info and it says I get 7.4mpg. Huge difference.
I'd also like to say, using your numbers, that Atheme isn't "closely following" Anope. Anope is used almost exactly 3 times as much or 300%.
Chaz, sorry for any confusion, I was simply commenting on the usefulness of using downloads as proof of anything. I wasn't accusing or implying that you had meant anything other than what you actually said.  |
|
| Back to top |
|
 |
nenolod Idler

Joined: 23 Jan 2004 Posts: 357 Location: A box!
|
Posted: Jul 16, 2011 11:06pm Post subject: |
|
|
i would say that atheme and anope have pretty much equal 'market penetration'. a lot of networks do not participate in searchirc.
in terms of 'networks on searchirc', which mainly is attractive to the unrealircd crowd, it is probably somewhere around 3.5:1 ratio for anope networks to atheme networks. that is an educated guess, mind.
a lot of the networks running atheme+charybdis simply don't participate in searchirc. there is about 400 networks running charybdis, based on the numbers i have come up with based on support inquiries in #charybdis.
so, realistically, i would say in the irc world, of which 'the networks which use searchirc' is a subset of variable size, i think atheme and anope are on pretty equal footing.
however, if you want to fudge the numbers a little bit, by usercount it is a little different.
freenode, espernet, darkmyst, and many smaller networks run variants of atheme = ~60-70k users served concurrently excluding duplicate users (e.g. on more than one aforementioned network)
rizon, swiftirc, sorcerynet and smaller networks running variants of anope = ~40k users served concurrently excluding duplicate users (there is a lot of crossover between rizon and swiftirc).
but really it is all statistics and meaningless as both projects target radically different philosophies of network administration... |
|
| Back to top |
|
 |
Trixar_za Eleet

Joined: 10 Dec 2006 Posts: 624 Location: South Africa
|
Posted: Jul 17, 2011 3:00am Post subject: |
|
|
So let's work with the REAL numbers: 5,538 servers as stated @ http://searchirc.com/networks
First, I'll sum up what I said before - this STILL holds true for this:
1.) I stated twice that it's a ROUGH estimated based on a SMALL (less than 50!) user sampling from the IRC Wiki.
2.) My final calculation INCLUDES the possibility of OTHER Services packages
Anyway, let's redo my math, shall we?
3/6 (0.5) for Anope
2/6 (0.333) for Other Services package
1/6 (0.167) for Atheme
Now:
5538 * 0.5 = 2769 for Anope
5538 * 0.167 = 925 for Atheme
5538 * 0.333 = 1844 for Other Services
All figures have been rounded.
NOTE: The original author of this thread asked for numbers pertaining to networks listed ON SearchIRC. He also did not state he required exact numbers. I provided him with rough estimates based on the information available to me. I.E. If you have a problem with it, please feel free to create your own figures...
PS: nenolod may be correct in his assumption that there is a 3:1 ratio compared with Anope vs Atheme usage. |
|
| Back to top |
|
 |
PingBad Post Whore

Joined: 05 Feb 2005 Posts: 3027 Location: New Zealand
|
Posted: Jul 17, 2011 4:38am Post subject: |
|
|
Statistics 101: larger, more diverse sample size = more accurate averages
That said, can you two not bicker? You've both raised a fair point but please leave it at that - just a personal request |
|
| Back to top |
|
 |
nenolod Idler

Joined: 23 Jan 2004 Posts: 357 Location: A box!
|
Posted: Jul 17, 2011 3:24pm Post subject: |
|
|
| Trixar_za wrote: | | 5538 * 0.333 = 1844 for Other Services |
i do not really think 1,844 networks are running services other than atheme or anope. that number seems really odd to me... |
|
| Back to top |
|
 |
katsklaw Guru

Joined: 28 Jun 2004 Posts: 1614 Location: Somewhere you're not.
|
Posted: Jul 17, 2011 7:23pm Post subject: |
|
|
PingBad, u first. According to the rules, we are allowed to say when we think ideas are stupid/false .. etc. Which is all I've done, someone voiced an opinion and I voiced mine in return. Nothing wrong with it.
Nenolod, yes, all the numbers are seeming odd. According to the Net Ranks page it's #N out of 855. 855 in my opinion is far more believable as far as active nets.
I'm truly interested in how you verify the exclusion of duplicate users, no sarcasm intended.
Trix, SIRC also claims 5,547 servers, so you are saying that's spread across 5538 networks?! That's an average of less than 1 server per net. Likewise the OP didn't ask for roughly calculated figures based on inaccurate numbers and guesses either. Which is exactly why I never bothered to give any "real" numbers, because I know for a fact that I'd only be guessing. The ONLY way to be accurate is for scrawl to log who uses what services and I seriously doubt that will happen, thus my "you can't" statement is still completely true.
My numbers, and they are more accurate than any given so far:
some use Atheme
some use Anope
some use other/none
Inactive nets aren't using services so they don't count. Non-indexed nets also don't count since the OP specified nets listed here.
Good Day everyone!  |
|
| Back to top |
|
 |
Trixar_za Eleet

Joined: 10 Dec 2006 Posts: 624 Location: South Africa
|
Posted: Jul 17, 2011 9:09pm Post subject: |
|
|
*sigh* My 'guesses', as you put it, can be tested. People can go count for themselves and draw their own conclusions (yes - even you katsklaw). I even gave you the figures and the math that I used. As PingBad has stated, the bigger the sample size, the easier it would be to determine more accurate stats.
Bitching how inaccurate my numbers are won't fix the fact that they do not exist. Nor does it change the fact that neither Atheme or Anope makes it easy to determine if the network is running either of them. All I can say is if you can do better then please do.
Also fine, let's use 855 as stated the network rankings:
143 = Atheme
428 = Anope
284 = Other
----------------------
855 Total Networks
Yes, these figures may be inaccurate, but hey, they exist now which is a lot better than guesstimates. |
|
| Back to top |
|
 |
|