A new website, and a clash in the hostel IP pattern
So yeah, the last post here was from January 2023. I did say I would try to keep this updated.
And I did not.
The site has been rebuilt from scratch. The old one was the Lagrange theme for Jekyll on GitHub Pages, set up in 2022 and not really touched since early 2023, and it still described me as an undergraduate. This one is a SvelteKit site running on Cloudflare Workers, at the same address. I rebuilt it with Claude, Anthropic’s AI model, working in Claude Code, which wrote most of the code and drafted this post while I directed what went where. Its structure follows what I actually do now: Research, Coursework, CV, Software, this blog, and the About and Contact pages. The colours are the nucleotide palette (A green, C blue, G amber, T red), and the track on the home page is a stylised genome browser, one gene with long reads supporting four isoforms, which is more or less what my PhD is about.
The three real posts from the old site are carried over. The old links redirect to their new places, including the blog posts’ dated URLs and the RSS feed, so anything that was shared around should still work.
The hostel IP page
The page of the old site that still gets shared around is the calculator for static IPs in C V Raman Hall, which I made in 2022 from CSSC’s note on the IP pattern for the boys’ hostel. Apparently it is the only place where the IP allocations still live, as CSSC’s new website does not have them. That link now redirects to the new hostel IP page, which covers Irene J Curie Hall and Meghnad Saha Hall as well. Pick your hostel, enter your room number, and it gives you the full set of settings with a copy button for each; once your connection is up, it shows that you are on the ISI Kolkata network (your IP is checked). CSSC’s notes for all three halls are linked at the bottom of the page.
Six pairs of rooms on one address
While porting it, I checked CSSC’s pattern against the actual list of rooms in C V Raman Hall, and it does not hold up. The note gives the last part of the address, 192.168.22.x, as
with bases of 20, 90, 130 and 170 for the North, East, West and South wings, and a step of 16 on North and South and 10 on East and West. The step only works if it is at least the number of rooms on a floor. That holds on North and South, which have 16 rooms a floor, but the East and West wings have 12 rooms on their second and third floors. Rooms 11 and 12 on the second floor therefore land on the addresses of rooms 1 and 2 on the third, and on the West wing’s third floor, on the South wing’s first two ground-floor rooms. The East wing’s third-floor rooms 11 and 12 happen to be safe, since the West wing has no ground floor to collide with.
| Rooms | Address under CSSC’s pattern |
|---|---|
| E-211 and E-301 | 192.168.22.121 |
| E-212 and E-302 | 192.168.22.122 |
| W-211 and W-301 | 192.168.22.161 |
| W-212 and W-302 | 192.168.22.162 |
| W-311 and S-001 | 192.168.22.171 |
| W-312 and S-002 | 192.168.22.172 |
I stayed in E-302, and used to have intermittent connection problems there that I never got to the bottom of. Under CSSC’s pattern, E-212 had the same address. The note’s own worked examples do not help either, since they use E-214 and S-324, neither of which exists.
For now, the page moves the six rooms that do not fit, the 11th and 12th on each East and West floor, to 192.168.22.85 through .90, which no room uses, and keeps everyone else on the address CSSC’s pattern gives them. This only works if both rooms in a pair follow the page, so if you are in any of these twelve rooms, the page tells you which room you clash with and what to do if your device reports a conflict. I have reported it to CSSC, and will update the page with whatever they decide.
If something on the site is broken, or your room’s address does not work, do email me.
Drafted with assistance from Claude Code.