1
00:00:00,270 --> 00:00:06,230
‫Obviously, great technology has a lot of applications, is it ready?

2
00:00:06,780 --> 00:00:08,040
‫Absolutely not.

3
00:00:08,550 --> 00:00:11,330
‫It's not ready for public consumption.

4
00:00:11,340 --> 00:00:12,730
‫I do not think it's ready.

5
00:00:13,440 --> 00:00:17,730
‫I think it has a great applications and needs a lot of perfection.

6
00:00:17,940 --> 00:00:20,880
‫There's a lot of work on the IBM team and other teams as well.

7
00:00:20,890 --> 00:00:26,940
‫I'm not sure what who, who what other companies are researching this this technology.

8
00:00:27,570 --> 00:00:34,440
‫But as of now, it's very, very, very slow because, you know, guys searching through 48 records

9
00:00:34,440 --> 00:00:36,870
‫takes a fraction of a nanosecond.

10
00:00:36,870 --> 00:00:39,860
‫Should write, should have been OK.

11
00:00:39,870 --> 00:00:41,280
‫That might be an exaggeration.

12
00:00:41,280 --> 00:00:42,440
‫A fraction of a second.

13
00:00:42,480 --> 00:00:44,340
‫This is not a speed of light kind of a thing.

14
00:00:44,640 --> 00:00:45,660
‫But you get my idea.

15
00:00:45,660 --> 00:00:46,280
‫You get my point.

16
00:00:46,560 --> 00:00:48,210
‫All right, guys, I'm going to leave it at that.

17
00:00:48,220 --> 00:00:50,240
‫Guys are pros and cons of this.

18
00:00:50,400 --> 00:00:53,040
‫What do you guys think about homo morphic?

19
00:00:53,760 --> 00:00:56,220
‫Do you think is going to is going to be a fruitful.

20
00:00:56,250 --> 00:00:57,640
‫Do you think is going to flop?

21
00:00:57,930 --> 00:01:02,960
‫Do you think it's not ready for production use, which I think it's not from what we're seeing here,

22
00:01:03,480 --> 00:01:07,170
‫but yeah, throw some CPU cores on it and see how it performs.

23
00:01:07,950 --> 00:01:10,620
‫Build a good index tearless termination.

24
00:01:11,010 --> 00:01:16,470
‫Well, probably I'm not going to send in my team and my reverse proxy or load balancer, to be honest,

25
00:01:16,470 --> 00:01:22,830
‫because, man, if I'm going to look at the stuff, imagine sending on a quest and takes a few minutes

26
00:01:22,830 --> 00:01:29,940
‫just to look and try try to make sense of this thing for recommendation engines, guys.

27
00:01:30,360 --> 00:01:31,290
‫Absolutely.

28
00:01:31,290 --> 00:01:31,950
‫Yes.

29
00:01:32,220 --> 00:01:40,110
‫So I think back to some of my what I said asynchronously, you can spin up some processes that does

30
00:01:40,680 --> 00:01:47,340
‫Homma Merfolk searching on encrypted data and generate like a Twitter recommendation to order Tranz.

31
00:01:47,340 --> 00:01:48,960
‫Who cares how long that takes?

32
00:01:49,080 --> 00:01:50,430
‫Yeah, the faster the better.

33
00:01:50,430 --> 00:01:52,770
‫But it's going to finish eventually.

34
00:01:53,130 --> 00:01:58,290
‫And I know no user is waiting for that.

35
00:01:58,740 --> 00:01:59,220
‫Right.

36
00:01:59,250 --> 00:02:07,020
‫It's not a synchronous operation.

37
00:02:07,140 --> 00:02:07,560
‫She is.

38
00:02:07,560 --> 00:02:08,400
‫How do you say that?

39
00:02:08,550 --> 00:02:11,280
‫It's not synchronous operation.

40
00:02:11,850 --> 00:02:12,140
‫Right.

41
00:02:12,150 --> 00:02:14,580
‫It's not like the users click and waiting for the result.

42
00:02:14,700 --> 00:02:18,450
‫It's not like the users tweeting and waiting for the tweet to be posted.

43
00:02:18,720 --> 00:02:22,040
‫Well, that's that's that's that should be just faster.

44
00:02:22,050 --> 00:02:24,060
‫I just just that's incorrect and right.

45
00:02:24,420 --> 00:02:27,900
‫The search the Twitter search obviously is going to be slow a little bit.

46
00:02:28,020 --> 00:02:30,180
‫Asynchronous search is going to be a little bit faster.