mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > FactorDB

Reply
 
Thread Tools
Old 2021-09-14, 15:47   #12
chris2be8
 
chris2be8's Avatar
 
Sep 2009

217410 Posts
Default

Flooding factordb with garbage entries is probably illegal in most countries that have laws about computer misuse. But only Syd could tell where the garbage is coming from or start legal action by complaining to the police. I wish he would.

Of course if someone starts bragging about flooding factordb by all means report them to the police.
chris2be8 is offline   Reply With Quote
Old 2021-09-15, 13:58   #13
Dr Sardonicus
 
Dr Sardonicus's Avatar
 
Feb 2017
Nowhere

497110 Posts
Default

IMO any submission to factordb of a "composite with no known factors" which is "too small" (fewer than 70 decimal digits, say) should be rejected out of hand. IMO those flooding, or trying to flood the site with such submissions should be summarily executed blocked.
Dr Sardonicus is offline   Reply With Quote
Old 2021-09-16, 02:06   #14
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

24·13·47 Posts
Default

Those things are not easy to do.

I am not a fan of Syd either (more specifically, his way to maintain things and his non-communicative person), as I mentioned in some previous posts (the authentication on that site still sucks, and I can login like almost anybody else, which still means nothing, as your work is not credited to you anyhow).

But to be objective, "flooding" of the DB is, most of the times (with only few exceptions of deliberate vandalism or stupidity), accidental. That is because how the DB works, if you look for formulas, it will try to generate the series for you. For example, one user tries "3n^2+69" or other stupidity which is very simple, but pick random constants with a low probability that somebody else tried in the past, and voila!, the DB is filled with over 3000 small composites, because it will try to generate them as high n as you look for, and all of them shatter in smaller pieces. Cutting off those features will be even worse (I use it myself many times, looking for CRUS sequences or aliquots, and it is indeed a VERY useful feature).

So, you would need a kind of very clever algorithm to parse your formulas and know when to fill the DB with all those things and when not. The "wrench" method is way easier to implement: the authentication and crediting the users properly, similar to what PrimeNet and thousands of other sites do (damn, no technical forums around, like stackexchange, etc, would let you post comments to other posts unless your "credit" is high enough, like 50 or so, which is not easy to get!), and allow only "trusted" users (i.e. those who do really work for the DB, like factoring, proving primes, etc) to create more than "n" indexes per "time unit" (n and t to be chosen by the admins). And we are back to my pain, haha, crediting system...

Last fiddled with by LaurV on 2021-09-16 at 02:15
LaurV is offline   Reply With Quote
Old 2021-09-16, 16:00   #15
chris2be8
 
chris2be8's Avatar
 
Sep 2009

2×1,087 Posts
Default

There are loads of entries like http://factordb.com/index.php?id=1100000002692465505
Code:
(10^256*2-1)*(10^256-1)*(10^240*9-1)*(10^240*6-1)*(10^240*2-1)*(10^240*8-1)*(10^240*3-1)*(10^240*7-1)*(10^240*5-1)*(10^240*4-1)*(10^240-1)*(10^224*2-1)*(10^224*3-1)*(10^224*8-1)*(10^224*4-1)*(10^224*6-1)*(10^224*5-1)*(10^224*7-1)*(10^224*9-1)*(10^224-1)*(10^208*8-1)*(10^208*6-1)*(10^208*9-1)*(10^208*4-1)*(10^208*7-1)*(10^208*2-1)*(10^208*5-1)*(10^208*3-1)*(10^208-1)*(10^192*6-1)*(10^192*2-1)*(10^192*3-1)*(10^192*4-1)*(10^192*5-1)*(10^192*7-1)*(10^192*8-1)*(10^192*9-1)*(10^192-1)*(10^176*2-1)*(10^176*3-1)*(10^176*5-1)*(10^176*4-1)*(10^176*6-1)*(10^176*8-1)*(10^176*9-1)*(10^176*7-1)*(10^176-1)*(10^160*7-1)*(10^160*5-1)*(10^160*3-1)*(10^160*8-1)*(10^160*2-1)*(10^160*4-1)*(10^160*6-1)*(10^160*9-1)*(10^160-1)*(10^144*8-1)*(10^144*7-1)*(10^144*9-1)*(10^144*6-1)*(10^144*5-1)*(10^144*3-1)*(10^144*2-1)*(10^144*4-1)*(10^144-1)*(10^128*8-1)*(10^128*7-1)*(10^128*6-1)*(10^128*9-1)*(10^128*4-1)*(10^128*5-1)*(10^128*2-1)*(10^128*3-1)*(10^128-1)*(10^112*2-1)*(10^112*3-1)*(10^112*4-1)*(10^112*5-1)*(10^112*6-1)*(10^112*7-1)*(10^112*8-1)*(10^112*9-1)*(10^112-1)*(10^96*4-1)*(10^96*5-1)*(10^96*2-1)*(10^96*3-1)*(10^96*8-1)*(10^96*9-1)*(10^96*6-1)*(10^96*7-1)*(10^96-1)*(10^80*5-1)*(10^80*3-1)*(10^80*4-1)*(10^80*2-1)*(10^80*6-1)*(10^80*7-1)*(10^80*8-1)*(10^80*9-1)*(10^80-1)*(10^64*4-1)*(10^64*2-1)*(10^64*6-1)*(10^64*3-1)*(10^64*7-1)*(10^64*9-1)*(10^64*8-1)*(10^64*5-1)*(10^64-1)*(10^48*5-1)*(10^48*3-1)*(10^48*4-1)*(10^48*7-1)*(10^48*2-1)*(10^48*6-1)*(10^48*8-1)*(10^48*9-1)*(10^48-1)*(10^32*8-1)*(10^32*2-1)*(10^32*4-1)*(10^32*5-1)*(10^32*3-1)*(10^32*6-1)*(10^32*7-1)*(10^32*9-1)*(10^32-1)*(10^16*2-1)*(10^16*9-1)*(10^16*8-1)*(10^16*7-1)*(10^16*4-1)*(10^16*3-1)*(10^16*5-1)*(10^16*6-1)*(10^16-1)/21128027404011231796679987089238683164672043293734347282046280057193578894183767468329620622544976128153932421099218866770930355449282470547961374258627439180859183715567849104301257593632410818381900850357635788687184944699005395305846125992782125646438379456783974431565945347773333727609172924231625208211073348271485590689601479896188418548048611004717961244263035178689788728093220777991652289635813714618985588702795302640770263683408799558625395552914001305755973519700612728726378790223516800452270166696383487690764601947182978363725409783610645988676885563368463566022779975430626504037923008352840277848048708195248393785202569233498062611501282107375250866615538851656557685739743650316737240235134070184036122044090690553835467078159298298814982749061154507518662758331480931/2686547288053565305126143066170755136583589822324669440338472283271083^3/310000000000000031^8/223232321^16/3^68/7^30/11^17
It was probably added as just the product of the small bits in brackets, then factordb searched for small factors and found the divisors.

Could something like that be added accidentally?
chris2be8 is offline   Reply With Quote
Old 2021-09-19, 12:01   #16
Happy5214
 
Happy5214's Avatar
 
"Alexander"
Nov 2008
The Alamo City

2×379 Posts
Default

We also have the formula "64^63%(2^63-1)*3969", which FactorDB has decided to store as not 1, not 2, but 3 different IDs, despite the formula resolving to a 4 digit number! These are now cluttering the unfactored composites list, since by the way it can't tell that 3^4 * 7^2 is a complete factorization.

Edit: It also displays those composites as 23 digits on the status pages. What?

Edit 2: Any guess who entered those numbers into the database?

Last fiddled with by Happy5214 on 2021-09-19 at 12:18 Reason: Who added them?
Happy5214 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Semiprime Squared Prime Test ¿ ONeil ONeil 8 2020-12-06 06:45
Semiprime and Prime Generator ONeil ONeil 2 2020-12-01 04:14
...factorial(n) plus/minus 1 is semiprime enzocreti enzocreti 3 2020-09-02 16:35
Semiprime counting danaj Computer Science & Computational Number Theory 12 2018-10-14 03:16
Factored my first SemiPrime -Some Questions tal Msieve 31 2011-05-22 18:17

All times are UTC. The time now is 12:58.


Sun Oct 17 12:58:32 UTC 2021 up 86 days, 7:27, 0 users, load averages: 2.61, 2.78, 2.55

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.