WHERE THIS IS
Nothing is running. This is the argument, written down first.
Segel is a design, not a service. There is no product to sign up for, no code published, and no round has ever been put through it. What is written here is the reasoning, set down before anything is built so that a buyer, a vendor, or an engineer can tell us where it is wrong.
What the design commits to
- 01A bid is binding the moment it is submitted, the same as returning a signed quotation.
- 02Before the deadline no bid is readable by the buyer, by the other bidders, or by us. Not restricted by a permission setting, but never delivered to a party who is not a stakeholder in it.
- 03At the closing time every bid becomes readable to the buyer and to all bidders in the same moment.
- 04After opening, a losing vendor can see the winning number. That is what makes a loss legible rather than suspicious.
What has to be built
- 01The sealed-round templates, and the disclosure choice that opens every bid at once. None of us has written Daml, so version one will miss things.
- 02A test suite proving no choice exists that opens a bid before the deadline. The claim above is worth nothing until that is checkable.
- 03The cases real tenders have: addenda that revise a specification mid-round, bidders who withdraw, extensions when too few bids arrive, split awards across two vendors. Each one changes what a bid contract must allow.
- 04An independent review before a real round is ever put through it.
What already exists
- 01Our own e-commerce platform, written and maintained here, serving buyers across Indonesia on servers we run.
- 02The operational side: upgrades, monitoring, and the restore you do at four in the morning because a disk filled up.
- 03This argument, in full, open to being contradicted.
If you run tenders or bid in them, the useful thing is not interest, it is a correction. Tell us the shape of a round you have actually run and where the design above would have broken.