Review by Peng Shuping

[msporny]

@shupingpeng provided the following review:

I read through your drafts on multibase and multhash. I got a few questions and comments.

1. Both drafts are Informational. Do you intend to put them as in the Informational track? Since it requires the IANA registries, should it be in the Standards track?

2. “Given data 'd' encoded into text 's', what base is it encoded with?”. You are talking about encoding bytes into text, but here you use ‘d’, what are the bytes ‘d’ is corresponding to? The example would be better to show using different base-encodings the same bytes could result in the same text, right? That could give your work much stronger motivation.

3. In the multihash draft, the “avoiding” part of this sentence could be removed. :p “This vastly simplifies the upgrade process, avoiding the waste of hundreds or thousands of software engineering hours, deep frustrations, and high blood pressure.”

Best Regards,

Shuping

[msporny]

On Mon, Mar 20, 2023 at 4:11 AM Pengshuping (Peng Shuping) pengshuping@huawei.com wrote:

I read through your drafts on multibase and multhash. I got a few questions and comments.

1.      Both drafts are Informational. Do you intend to put them as in the Informational track? Since it requires the IANA registries, should it be in the Standards track?

The outcome of the Multiformats dispatch discussion was to determine if an ISE path, or a mini-WG path, is best. The emails over the next couple of weeks will determine the path that the group will take. I will note that there was some disagreement on whether or not the ISE path could establish a registry, which is largely what the Multiformats specifications are about.

We'll see if we can get a firm response to that question over the next couple of weeks.

2.      “Given data 'd' encoded into text 's', what base is it encoded with?”. You are talking about encoding bytes into text, but here you use ‘d’, what are the bytes ‘d’ is corresponding to?

'd' refers to any arbitrary bytes. The base that the text is encoded with is expressed as a Multibase value at the beginning of string 's'.

 The example would be better to show using different base-encodings the same bytes could result in the same text, right? That could give your work much stronger motivation.

There are a number of examples in the specification today that show a single value being expressed in different base encodings here:

https://www.ietf.org/archive/id/draft-multiformats-multibase-07.html#name-test-values

... but what I believe what you're asking for is an example in the introduction of the specification to demonstrate why having a header in string 's' is desirable. Without that header, you won't know what base encoding the string should use. Did I understand your request correctly?

3.      In the multihash draft, the “avoiding” part of this sentence could be removed. :p “This vastly simplifies the upgrade process, avoiding the waste of hundreds or thousands of software engineering hours, deep frustrations, and high blood pressure.”

While I believe the grammar in that sentence is correct, I'll try to simplify it so that it conveys the point in a better way. The sentence is a bit hyperbolic, and so clearly articulating the benefits might be more appropriate.