1410 k let's make multithreading work 1431 return self.peer_stream(txid, range=range) File "/home/ubuntu/src/pyarweave/ar/peer.py", line 1062, in peer_stream return io.BufferedReader(PeerStream.from_txid(self, txid), 0x40000) File "/home/ubuntu/src/pyarweave/ar/stream.py", line 12, in from_txid tx_offset = ar.Transaction.frombytes(peer.unconfirmed_tx2(txid)).offset AttributeError: 'Transaction' object has no attribute 'offset' i'll just spend some time learning how to get the offset of a transaction from its binary object. looks like a bug in my pyarweave library. very inhibited, so the task change is actually helpful. 1431 boss 1: "why are u vomiting everywhere" boss 2: "i ate an ai i'm so sorry please excuse me" [vomits left] [vomits right] boss 1: "dude we have messed up software developers to do that for you" boss 2: [vomits] 1433 1436 i'll have fun checking out the function in arweave that calculates the offset 1439 https://github.com/ArweaveTeam/arweave/blob/master/apps/arweave/src/ar_data_... i wonder if i've reviewed this before 1629 just got the same checksum from a multithreaded upload, as download. major threshold :) met my hard subgoal. met it !! 1631 i'm going to use my goal coping stating of mind to try to add more features, like gps and photo capturing, and system logs, while recording is running. 1640 thinking about how much it would help to analyse problems to have blockchained logs. thinking about it stimulates lots of inhibition ^_^ 1641 1657 k i'm confused around this; i'll take it slowly. previously the generator produced a sequence of data chunks and the storer uploaded each one now each one has a type. some of these make sense to upload, some make sense to include raw. it's no longer a list of things to just pop and upload: one needs now to consider the order in which they are uploaded. when they are indexed, the indexer has to wait for stuff, so it can put a few in concatenated. it makes sense to separate the types out for the concatenation: accumulate separately for each type. if the types are separated out, this would happen after storing, since storing operates in parallel. it would make sense for the concatenation to happen at the end of storing. 1706 ok the storers retain an index in order to keep things in the right order while operating in multiple threads. ummmmm atm in the implementation, _all_ the events are ordered. ok. hum. 1714 to handle multiple binary streams i'll need to add channels to flat_tree . leaves two unstable things at once, so my plan is to have only 1 binary stream for now. 1715 actually i'm fudging it so it only tracks the length of one stream, but lets them all be there 1720 doing my goal-focus thing as stuff pushes away from my task 1747 there's a bug, it's associated with disorganization in my code. i've seen it before. i feel frustrated, exhasperated, unneccessarily. the small weakness opens big space for inhibition. there are a few things going on at once. i'll focus on a different bug immediately. might not be the best choice. 1757 so what i seem to settle into is a state of mind where i have an extended sense of what i call 'inhibition'. it's a set of intense experiences that i usually respond to by leaving my current task. what's interesting, is that i'm in a state of mind where i can keep working some. it's hard, confusing, but happens. i guess it built a little out of the 'dissociated' working, where at base i would just copy code back and forth. it's really unpleasant and i'd like to find something different or a way to improve it. it is of course great that i can continue to do some work. 1758 i'm thinking about it a little bit, and i can kind of feel in me that there are some ways i might improve it, similar to how i started this. thinking of just doing a little bit, and getting better at that. i could pair it with some distance. there are inner spaces that are hard to describe, but maybe it's really reasonable, unsure. 1759. 1805 - the extra 'None' output appeared to be just a wrapping line of debug output. helps to turn off debug output for dependency. - ... presently 3MB turned into 1.7MB, likely truncated. 2106 i've observed a repeating race condition in the multithreaded uploader that produces an unending hang. this is only happenign with the new code, but occurs in the old code. it looks to me like the right solution is to change the layout, which is more work than i planned for. spending some time thinking about how things could be easier for me. not sure what to think on this, but it seems like what makes sense to consider. getting up in the morning is so hard. physically and cognitively. it's like a puzzle, finding parts that will do it. i've been finding them a little more readily, thinking of the strategies i came up with from the second chapter of that book, but it's not the be all and end all -- i have dissociation and dyskinesia and amnesia and other similar stuff, that can really make things confusing and complex. some of the behaviors i've built over the past two days work, but they don't feel right. the dissociation associated with them is very palpable. i have a state of mind where i'm working that i'd like to protect better. similar to where i'm getting up. but i guess getting my parts able to do these things is a step towards a solution. thinking of addressing this enough to get through much of it is such an incredibly different viewpoint from years past. there is a lot that goes undescribed. lots of important things. 2110 the goal is to reduce the inhibition. sleeping is not my favorite thing all the time. my inhibitions can sometimes seem to strengthen a lot. maybe there are things to think about around that. the goal is to reduce the inhibition. we can also skill build around goals and behaviors, but it seems stronger sometimes to reduce the inhibition. 2113 2115 tomorrow is tuesday the 16th. i'm working on moving some code inside multicapture.py, so as to accommodate when a network exception is thrown after popping an item off the queue to process