...cont....
I thought the street topologies survey was a great idea! But it missed the mark because it was way to in-depth for a layperson to understand and I thought people would be turned off by it thus greatly skewing the end results of the survey. I'm not an urban planner or a civil engineer but just someone who's always found urban planning interesting and thought I could maybe make a difference. Feels good to be heard. Just wanted to share :)
TL;DR - I read this book, and found it very useful in solving some organizational / management problems I was encountering.
https://preview.redd.it/xpwxe0f3wi881.png?width=1122&format=png&auto=webp&s=df1dcfa2cc2f7c2c64e44bd210f1afc17a7c365b
Problem I was solving: Bridging some gaps in my knowledge regarding how to design and structure teams in larger, complicated programs with lots of stakeholders, especially with government clientele. Also, I was looking for some clear frameworks to describe some issues some of my product teams were having: Product people on one particular team reporting that they didn't feel like they were equipped to meet the demands placed on them for a number of reasons, and that they and some others were starting to feel burnt out
Context: I'm a Product Management Leader at a design and engineering services firm that builds software for CMS (Centers for Medicaid and Medicare Services). I have ~10 YOE in healthcare product management, all in startups or growth-phase companies, with ~5 YOE in engineering before that. I've read and applied most of the more popular design thinking / product management books, e.g. Marty Cagan's books, Service Design, etc.
How it helped me, and why you might want to read it: The book provided a clear framework grounded in Conway's Law for structuring teams in an organization to meet the needs of users and align with the architecture of the systems you're building (whether they're services or software or hardware). It put product teams into the context of a larger, more flexible framework that covers more real-world circumstances that product leadership may encounter in large systems and programs. There's probably a better summary, but I now have a very useful framework to typify teams and the modes of interaction they use to communicate, and I can leverage that framework and a number of best practices in my recommendations for improving the way we fulfill our contracts and meet the needs of users, or even informing how we should bid on contracts.
Who I would recommend read the book (how much value I think you'll get out of it):
- 10/10 - Directors of Product or Product Management practice
- 10/10 - Leadership of product-led organizations (e.g. CTO, Dir. of Engineering, Dir. of HCD, etc)
- 7/10 - Senior-level Product People with some influence on team
I'm trying to create a network topology, utilising existing tools. It's an international network infrastructure, so far I've used mtr for traceroutes, wrote a shell script to use freegeoIP to obtain longitude & latitude, then used ggplot2 on R to plot this data. (The idea is for this to be used to discover connections to new domains without knowing the connections beforehand)
But it's been very difficult to understand the topology, especially trying to dumb it down to the executives. So I used zenmap/nmap to visualise the topology. That worked really well! But now it's missing the geo reference.
I've also tried visualising it with gephi - which worked great! But again, at the expense of geo reference points.
Open to non FOSS tools, as long as it can overlay the topology on a national/global map.
[EDIT]
I've also tried this:https://github.com/BugenZhao/flashroute.rs Which is insanely fast! Could be useful for other use cases.
My POV is of a physicist interested in the mathematical foundations of GR (and other metric theories of gravity) and whose knowledge of topology and real analysis is all self-taught and patchy. So I suppose "point set topology" is my only interest here, and that's what I'm asking about. I understand that pure mathematicians don't need a reason, they just like having algebraic structures to poke and prod, so if you're a pure mathematician just pretend you care about applications for a second.
I first was introduced to it as the study of "continuity of maps", and I learned how the topology 101 definition of a continuous map maps (heh) exactly to the epsilon-delta definition of continuity in real analysis. Then I went digging through real analysis and topology books and I think I've pieced together the following applied mathematician's/physicist's motivation of topology:
>With metric spaces we study the continuity of maps (and hence differentiability, which we need for physics) using a generic definition of "distance" (not necessarily Euclidean distance, or a vector inner product, or something like that). Topology is the study of continuity of maps at its most fundamental, i.e. without needing to invoke a concept of "distance", so that continuity can be studied in contexts more general that functions from R to R. In this way we can study differentiability with the absolute minimum of assumptions and extra structure.
Okay that all sounds fine, but it seems to me that any topological space we'd want to study would have a metric defined on it, and the open sets that make up a topology on a given set are usually chosen to be workable with a (generic) metric. And so it seems to me that the "minimal assumptions and minimal structure" thing is a bit misleading since we're really choosing open sets (or bases for them) with (generic) metrics in mind. So we've kind of built the metric space structure into our topologies with our choices of open sets. Or if not the full structure, the socket that it plugs into.
I'm worried that if (given a set) we constructed the open sets of a topology for it in some way not amenable to a metric space, then for that topology we'd have maps that would fit the definition of "continuous" but wouldn't be what anyone would actually call continuous, if presented with the map in isolation, but right now this is just a vague idea in my head that I haven't pinned down.
In which case, why not just talk about metric spaces all the
... keep reading on reddit ➡Right of the bat, this thing feels like a premium object. It's heavy, it's made extremely well and looks great. The Delrip wrap on it keeps it from getting ice cold in the winter. That said, I'm coming from the STI leather-wrapped shift knob and wasn't aware of how much I liked the grip on it. The Delrin wrap has no grip it all. I think I was expecting something with just a bit more texture or grip. That said, I can still shift fine, but it slides around in my hand when I do. But again, nothing wrong with it overall; it's a really well-made shift knob, but if grip is important to you, you might want to consider something else.
https://preview.redd.it/0kxz9ah5woa81.jpg?width=4032&format=pjpg&auto=webp&s=aec602995513c40747046be09cdf93f87289cb55
I'm trying to ascertain the differences between an education at a general European university vs an American one. I've heard phrases like "first year analysis" or "first year topology" before but those seems like pretty advanced classes for a new/freshman student to take and am wondering if that's what is expected of you
edit: Thank you all for all of the replies. They were very informative + helpful
Hi All,
A new factory is being built, about 1200mx600m in size (the building itself), lot of equipment inside/outside the factory will require network connections. I will be responsible for IT, but for now I'm just doing a consultancy job for them, because it's only 20 people there or so. It's all just started. We've finally received the full technical project of the factory, including computer networks (no equipment, just cables, cabinets location and general idea). The engineering company who made this project proposed the following design (I redrew it in Visio to exclude unnecessary details/foreign language). The distance between cabinets is in meters.
Project company proposed design: https://ibb.co/WymySfx
basically it's a ring (they call it "star" in documentation ;) that terminates in the main comms room of administrative building. According to documentation 24 strand multimode cable will be used.
I see multiple issues in ring topology like this and would like to come up with classic star topology and core/distribution/access tiers):
redesigned : https://ibb.co/ChvBRSq
It's a factory, not a media company with heavy traffic. End clients will be AP, computers, factory equipment, cameras, sensors etc.
I have a good networking background, although it's not my main skill for sure. I'm afraid being missing something. Single/multimode and SFP costs is my concern too. comparing with all network equipment and other stuff it's probably nothing, but again - maybe I'm missing something. Obviously don't want to look silly raising my concerns to the engineer company, asking them to change the topology.
So my pro and contra of both designs:
Ring
Advantages:
• Less fiber cable/length required - cost saving
• Cheaper SFT fiber modules (multimode fiber) - cost saving
Disadvantages:
• Full network throughput is equal to one slowest switch throughput. Because the whole network/all switches sharing the same single bus.
• Network upgrade requires all components upgrade
• Any two parts (switch, optic, sfp) failed will bring whole network down. Ring topology cannot survive more than one link loss.
• Any part failure (switch, cable, port, module etc.), maintenance or just a reconfiguration may cause STP route rebuild which may end up in temporary packet lost, may impact production. You always must be extremely careful with any change.
• Very hard to diagnose issues, because you cannot isolate anything
... keep reading on reddit ➡I'm trying to decide on a sequence of math courses to take over the next couple years. I am interested in both of these subjects so my decision to choose which to work towards will largely depend on which will be more useful for what I want to apply them to which is machine learning or an kind of AI.
This is the description to the topology course that I would be working up to after taking intro to analysis, intro ring theory and intro group theory:
>General point-set topology. Compactness, Tychonoff’s theorem, connectedness. Metric spaces, completeness, Baire’s theorem. Urysohn’s lemma. Topological manifolds. Homotopy theory, fundamental group, covering spaces
And this is the description to the differential geometry course that I would be working up to after taking a differential geometry of curves and surfaces course and multivariable calculus
>Riemannian geometry of n-space, metric tensors, various curvature concepts and their relationships, covariant differentiation, geodesics, parallel transport. Additional topics at the discretion of the instructor
What would be the more useful subject to study in undergrad if I want to apply this kind of math to machine learning? Would knowledge in one of these subjects transfer over to the other subject or not really?
Thanks in advance for any advice.
Let S be an infinite subset of R such that S\{x} is compact for some x \in S. Then which of the following is true?
- S is connected set
- S contains no limit points
- S is a union of open interval
- Every sequence in S has sub-sequence converging to an element in S.
I think option 4 is right. The reason being:
If S\{x} is compact then its is closed and bounded. That implies S is also compact. x \in S can not be a limit point other wise after removal of x , S\{x} would not be closed[a closed set contains all its limit points], which would imply S\{z} is not compact. Now since S is closed and bounded, every sequence in S has sub-sequence converging to an element in S(property of closed set).
Am I right? Thanks in advance
Two professors are giving a student who isn’t very bright an oral exam. They ask “give an example of a compact topological space”, to which the student replies “the real numbers”. There is a long painful silence before one of the professors helpfully asks “in what topology?”
)