My money is still on some kind of Unix ephoc overflow.
Well, since Unix timestamps are signed
integers (which I didn't know until reading up on this), they can have negative values after overflowing, meaning that they actually can
(in a crude manner) represent dates in the year 1926. But if his posts are really from 2235, they would have had to overflow quite a few times.
Here are the dates together with the negative, overflowed timestamps that the forum uses (you can get these, when quoting one of Aadarsh's posts) and with the timestamps for the presumed post dates on Aadarsh's calendar (calculated with http://www.csgnetwork.com/time2unixdscalc.html
10/10/26 17:07:28 Can anyone hear me -1364086352 11520810448 Jan 30 2335 (16 days before next post => news report that peace with Vasudans is because of Shivan attack)
26/10/26 18:54:58 Not as it appears -1362697502 11522199298 Feb. 15 2335 (4 days before next post, "uploaded on the 15th" => news talk about HoL as Rogue Vasucan faction)
30/10/26 18:18:24 Interesting Times -1362354096 11522542704 Feb. 19 2335 (6 days before next post)
05/11/26 02:34:55 Some more Highlights -1361892305 11523004495 Feb. 25 2335 (4 days before next post)
09/11/26 11:02:32 On the destruction of a Multidude -1361516248 11523380552 March 1 2335 (news about finished destruction Vasuda => 3 news entries: Vasuda destroyed; Bastion helps with evacuation; Shivan assault continues)
16/11/26 01:39:14 Always one Step behind -1360945246 11523951554 March 8 2335 (7 days after prev. post)
07/12/26 11:42:23 Analysis and speculation -1359094657 11525802143 March 29 2335 (21 days after prev post)
Is there anyone who sees a pattern or who at least can confirm the correctness of these timestamps?