8x8 Inc (EGHT) Quote

does drive os send realtime data to the learning module during the course of operation? answer from geminiai




Yes, absolutely! DRIVE OS is designed to send real-time data to the learning module during operation. This is a crucial aspect of how autonomous vehicles (AVs) learn and improve their driving capabilities.  

Here's a breakdown of how it works:

DRIVE OS and its Components

Operating System: DRIVE OS is NVIDIA's operating system specifically designed for autonomous vehicles. It provides a platform for running the complex software and AI models required for self-driving.  
Sensor Input: AVs are equipped with various sensors, including cameras, lidar, radar, and ultrasonic sensors, which constantly gather data about the vehicle's surroundings.  
Perception: DRIVE OS processes the raw sensor data to perceive the environment, identifying objects, lanes, traffic signals, and other relevant information.  
Planning and Control: Based on the perceived environment, DRIVE OS plans the vehicle's path and controls its actions, such as steering, acceleration, and braking.  
Learning Module: The learning module is responsible for training and improving the AI models that power the AV's perception, planning, and control systems.
Real-Time Data Flow

Sensor Data Acquisition: Sensors continuously collect data about the environment.  
Data Processing: DRIVE OS processes the raw sensor data in real-time, extracting relevant features and information.  
Data Transmission: This processed data is transmitted to the learning module, which might be located on the vehicle itself or in the cloud.
Model Training and Updates: The learning module uses this real-time data to train and update the AI models, improving their accuracy and performance.
Deployment of Updated Models: The updated models are then deployed back to the vehicle, enhancing its driving capabilities.
Types of Data Sent to the Learning Module

Camera Images: Images from the vehicle's cameras, along with annotations or labels identifying objects and features.
Lidar Point Clouds: 3D point cloud data from lidar sensors, representing the shape and distance of objects.  
Radar and Ultrasonic Data: Information about the speed and distance of objects detected by radar and ultrasonic sensors.
Vehicle Data: Data from the vehicle's internal sensors, such as speed, acceleration, steering angle, and GPS location.  
Driver Inputs: In some cases, data about the driver's actions, such as steering corrections or braking, might also be sent to the learning module.
Benefits of Real-Time Data

Continuous Learning: Allows the AV to continuously learn and adapt to new situations and environments.  
Improved Performance: Real-time data helps improve the accuracy and performance of the AI models, leading to safer and more reliable autonomous driving.  
Faster Development: Accelerates the development and validation of AV systems by providing a constant stream of real-world data.  
Overall

The ability of DRIVE OS to send real-time data to the learning module is a crucial aspect of its functionality. This enables continuous learning, improved performance, and faster development of autonomous driving systems. By leveraging real-world data, AVs can become more intelligent, adaptable, and safe over time.  


Sources and related content

for the combined classical and quantum network to do the job of generative ai,
unified messaging/communications has to be maintained throughout the defined network, such as msft fabric. api/instructions/communications/answers/questions /parts of algorithms going to different parts involved in the parallel processing of the query need to be orchestrated and done is a reliable way so that the probability of their validity is obtained in the most reliable manner possible and in the time frame needed/desired. guesses.


timing, time, clocks and a multitude of clocks might be involved , just one example is qite, or more simpler word processing apps, especially, if any latency of their parallel work is encountered, might need a file system for storing the work until 'done'..guesses.

so, the different components of the system/architecture need to be ready to receive event driven messaging, data;perhaps, notified by sms?
and their partial work placed into a file until completed and then the app notifies the integrator, next step.

from another company:"Method and system for automatically downloading and storing markup language documents into a folder based data structure
Patent number: 9405736" from justia.

patent application 2013 and grant 2016....long live this patent, which would seem crucial to cuda-q, etc.
8 could use it to interoperate with the various apps and their timing?

"Focus: Fabric focuses on providing a unified platform for data and analytics, while .NET is a general-purpose development platform for building various types of applications.  
Deployment: Fabric is a cloud-based SaaS offering, while .NET applications can be deployed on-premises, in the cloud, or in hybrid environments.  
Technology Stack: Fabric leverages various technologies within the Azure ecosystem, while .NET is a specific framework for building applications on the Microsoft platform."

one platform for unified communications is 8 mantra.....

this .NET is a specific framework for building applications on the msft platform.
now, is this an arbitrary object platform in the manner of vcsy or borland delphi, which is willy nilly? the tools for building might say msft on them, but are they manifestations of v ip patents?

the key word for the presentation is do they use arbitrary in the manner of vcsy or pure chance.......from my readings on quantum stuff, there is no pure chance . everything is by design/rules of order/probabilities......

is microsoft fabric vcsy ips?
i asked gemini ai 1.5 pro :"is microsoft fabric a new .net platform?




No, Microsoft Fabric is not a new .NET platform. It's actually a unified analytics platform that brings together various data and analytics tools into a single, integrated environment.  

Here's a breakdown of what Microsoft Fabric is and how it relates to .NET:

Microsoft Fabric:

Unified Analytics Platform: Fabric combines components from Azure Synapse Analytics, Azure Data Factory, Power BI, and other services into a single platform. It provides a comprehensive suite of tools for data integration, engineering, warehousing, science, real-time analytics, and business intelligence.  
SaaS Offering: Fabric is delivered as a Software as a Service (SaaS) solution, meaning it's hosted and managed by Microsoft in the cloud.  
Key Features:
OneLake: A multi-cloud data lake that provides a centralized repository for all your data.  
Data Factory: A data integration service for creating ETL (Extract, Transform, Load) pipelines.  
Synapse Analytics: A platform for big data analytics, data warehousing, and data science.  
Power BI: A business intelligence tool for data visualization and reporting.  
Target Users: Data engineers, data scientists, data analysts, and business users.  
Relationship to .NET:

Not a .NET Platform: Fabric is not built on or directly related to the .NET platform. It's a cloud-based SaaS offering that leverages various technologies and services within the Microsoft Azure ecosystem.  
Potential for Integration: While Fabric itself is not a .NET platform, it's possible to integrate .NET applications with Fabric. For example, you could use .NET to build custom data connectors, process data, or interact with Fabric APIs.
.NET for Data Science: .NET can be used for data science tasks within the Fabric environment, particularly with tools like ML.NET.
Key Differences:

Focus: Fabric focuses on providing a unified platform for data and analytics, while .NET is a general-purpose development platform for building various types of applications.  
Deployment: Fabric is a cloud-based SaaS offering, while .NET applications can be deployed on-premises, in the cloud, or in hybrid environments.  
Technology Stack: Fabric leverages various technologies within the Azure ecosystem, while .NET is a specific framework for building applications on the Microsoft platform.
In Summary:

Microsoft Fabric is a powerful analytics platform that provides a comprehensive suite of tools for data professionals and business users. While it's not directly related to the .NET platform, there are opportunities for integration and using .NET for specific tasks within the Fabric environment. However, it's important to understand that Fabric is a distinct offering with its own focus and capabilities.  


Sources and related content
"

does microfrontend carry it's own time stamp?




It's not quite accurate to say that a microfrontend itself carries its own timestamp. However, timestamps can play a crucial role in managing and updating microfrontends within a larger web application.

Here's how timestamps relate to microfrontends:

1. Independent Deployments:

Microfrontends are designed to be independently deployable. This means that each microfrontend can be updated and released without affecting other parts of the application.  
Timestamps can be used to track the version and deployment time of each microfrontend. This helps ensure that the latest version of each microfrontend is loaded and displayed in the application.
2. Cache Busting:

Browsers often cache static assets (like JavaScript and CSS files) to improve performance. However, this can cause problems when a microfrontend is updated, as the browser might still load the old cached version.  
Timestamps can be used as query parameters in the URLs of microfrontend assets. This forces the browser to re-fetch the assets from the server, ensuring that the latest version is loaded.
Example: https://my-app.com/microfrontend-1.js?t=1678886400 (where t is a timestamp).
3. Version Control and Rollbacks:

Timestamps can be used in conjunction with version control systems to track changes to microfrontends and enable rollbacks to previous versions if necessary.
This helps maintain the stability and reliability of the application, especially when multiple teams are working on different microfrontends concurrently.
4. Dependency Management:

Microfrontends often have dependencies on shared libraries or resources.  
Timestamps can be used to track the versions of these dependencies and ensure that compatible versions are loaded.
5. Build Processes:

Timestamps can be embedded into the build artifacts of microfrontends during the build process. This helps track the build time and identify potential issues related to specific builds.
Important Considerations:

Timestamp Granularity: The granularity of timestamps (e.g., seconds, milliseconds) should be chosen based on the frequency of updates and the desired level of precision.
Synchronization: In a distributed microfrontend architecture, ensuring that timestamps are synchronized across different servers and clients can be important.
Alternative Versioning Strategies: While timestamps can be useful, other versioning strategies, such as semantic versioning or content hashing, might be more robust and reliable in certain situations.
In Summary:

While microfrontends themselves don't inherently carry timestamps, timestamps can be a valuable tool for managing their deployment, versioning, and integration within a larger application. By using timestamps effectively, developers can ensure that the latest versions of microfrontends are loaded, manage dependencies, and maintain the stability of the application.


Sources and related content

Device drivers interact with the kernel primarily through a well-defined set of interfaces and mechanisms:

Kernel APIs and System Calls:
Drivers use kernel-provided functions (APIs) to access kernel resources, manage memory, schedule tasks, and communicate with other parts of the system.
Drivers also implement specific entry points (system calls) that the kernel invokes to perform device-specific operations requested by user applications or other parts of the system.  
Interrupts and DMA:
Drivers use interrupts to signal the kernel when a device requires attention, such as when data is ready to be transferred or an error has occurred.  
Drivers may also use Direct Memory Access (DMA) to transfer data directly between a device and memory without involving the CPU, improving performance.  
Device Registration and Management:
Drivers register themselves with the kernel during system initialization, providing information about the devices they support and the operations they can perform.  
The kernel maintains a registry of loaded drivers and handles device discovery, configuration, and resource allocation.
Kernel Modules:
In many modern operating systems, drivers are implemented as loadable kernel modules, allowing them to be added or removed from the running kernel without requiring a system reboot.

the role of cpaas in generative ai


Show drafts


CPaaS (Communications Platform as a Service) plays a crucial role in enabling and enhancing the integration of generative AI into communication channels.

Here's how:  

Seamless Integration: CPaaS platforms provide the necessary infrastructure and APIs to easily integrate generative AI models into various communication channels, such as voice, SMS, chat, and video. This allows businesses to leverage the power of AI to enhance customer interactions, automate processes, and personalize experiences.  

Real-Time Communication: CPaaS enables real-time communication between generative AI models and customers, allowing for instant responses and dynamic conversations. This is essential for applications like AI-powered chatbots, virtual assistants, and voice assistants.  

Scalability: CPaaS platforms are designed to be scalable, allowing businesses to easily add or remove communication channels and AI capabilities as needed. This ensures that the generative AI-powered communication solution can adapt to changing business needs and growing customer demand.  

Omnichannel Experience: CPaaS enables businesses to provide a consistent and seamless omnichannel experience for their customers. Generative AI models can be integrated across multiple channels, allowing customers to interact with the same AI assistant across different platforms, such as website chat, SMS, or voice calls.  

Data Collection and Analysis: CPaaS platforms can collect and analyze data from customer interactions with generative AI models. This data can be used to train and improve the AI models, personalize customer experiences, and gain insights into customer behavior and preferences.

Security and Compliance: CPaaS platforms often provide built-in security features and compliance measures to protect customer data and ensure privacy. This is crucial when dealing with sensitive information that may be shared during interactions with generative AI models.  

Use Cases of CPaaS in Generative AI:

AI-Powered Chatbots: CPaaS enables businesses to deploy AI-powered chatbots that can handle customer inquiries, provide support, and even complete transactions.  
Virtual Assistants: CPaaS can be used to create virtual assistants that can schedule appointments, send reminders, and perform other tasks for customers.
Voice Assistants: CPaaS allows businesses to develop voice assistants that can answer questions, provide directions, and control smart home devices.
Personalized Marketing: Generative AI models integrated with CPaaS can create personalized marketing messages and recommendations based on customer data and preferences.
Conclusion:

CPaaS plays a critical role in bridging the gap between generative AI and communication channels. By providing the necessary infrastructure, APIs, and features, CPaaS enables businesses to leverage the power of AI to transform customer interactions, improve efficiency, and drive growth.


Sources and related content

shorting of rng, 8 , interesting. is 8 in a reversal position with rng?

211.226shares traded at 2.706 @12:42

covering or buying?

nasdaq says institutions own 102.81%
yahoo says institutions own 93.89%

a difference of 8.92% of 125 million shares.

i ask myself is the difference the number of naked short shares?11.150 million shares?

and a recent sec ruling made the market maker which allowed this transactions to make good on the naked short positions..

new chromebook plus, like acer's, allows access to gemini advanced. and integration google apps like gmail and docs.

rag might require obtaining data from google, meta, twitter, ihub, cnn, etc. does 8 have a license from such? they seem to have something with the google contact center in the past.

from the quarterly:"These are the early indicators that our efforts to build a high-performance go-to-market organization are working. Given the nature of enterprise sales cycles and our ratable revenue growth, there's a lag before our progress is evident in our revenue growth. But I am confident that the changes we've made will drive improved performance and accelerating growth by the end of the current fiscal year. As we look to the future, our strategy is clear. We will continue to drive innovation, intentionally engineered solutions that deliver superior business outcomes for our customers. I encourage you to watch a short video case study of the San Diego Zoo's implementation of our integrated solution as an example of our solution approach and rapid time to value.

Working in partnership with the internal team at the zoo, we implemented a full service contact center with an AI-based self-service option for inbound calls in less than 45 days. We have intentionally engineered our products and solutions for this level of agility and rapid time to value. We can leverage this advantage to develop our own purpose-built solutions as well. For example, 2024 is a big year for elections in both the US and UK. Using the power of conversational AI and automation, our UK team built a library of out-of-the-box templates to support local government and improve citizen access to important information for upcoming elections. To drive awareness and adoptions of our innovations, we are committed to continued improvement in our go-to-market activities."

the voter ai assistant and need for rag/retrieval-augmented generative for the llm , wags. sam mentioned rag in the quarterly.

rng down this am, 8 so far steady.....? reversal of the long/short going on. closing the shorts and going long on 8?june 1, a new day for 8?

personal assistant for voting information. so who provides the governance on the info?
today show just tried to say inflation is not so bad citing mcdonalds prices only up 21% 2019 to today. how much was in the last 3.5 years?

yahoo seems to have deleted my posts. but i guess investorhub is freedom from tyranny and manipulation.

8 and aws;
sap and aws for multiple chat apps in an app? sap and aws pr today.
ec2 of aws allows users to rent virtual servers . to use such and the services they provide you might also need to use the 8 patent for user designated dynamic dns mapping of virtua server and virtual services desired.
a review of the sap/aws pr shows there are 7 ai's available : AIw1 labs, anthropic, cohere,meta, mistral ai, stabiity ai, and amazon.

i'm wondering if apple will use some aws for their ai. i believe apple used ibm, which was on vz, which is now on aws.
small world...

surprised not any postings here in the past Month plus, it has been a steady Trader, No dilution.

EGHT................https://stockcharts.com/h-sc/ui?s=EGHT&p=W&b=5&g=0&id=p86431144783

EGHT.......................https://stockcharts.com/h-sc/ui?s=EGHT&p=W&b=5&g=0&id=p86431144783

Doc What we have here is a failure to communicate. Vcsy wants to communicate with eght. But they’ve got to get their minds right. AI. Get your mind right.

Hey Doc. Go eight !!!!I got on here. Only allowed 1 post per day over there. U say 1 wrong word and their ai knows and blocks or bans you. Be careful . Looks like I’ll be buying some eght ! Daughter looking at Oxford Lawyer/Judge/C J SCOTUS so yeah. I’ll need to buy my own G3. That’ll do. Lol

chrome and firefox browsers
cover android/hp/acer/samsung/lg/etc and apple products. wags.

chat apps and programmable apis......so various chat applications from major os/platforms, programmable integration of realtime streaming audio and video objects in javascript /webrtc(server side) to be treated like standard video/audio objects in html5( client side).

how to get the generative ai into your application.......so if an amazon provides generative ai, how do you get it into your application?
gosh, if this is so, this is huge. no wonder they've got these messages blocked.

self service push for improved customer eXperience, including costs savings?
segue with amazon web service work of customizing for customers. wags.
using programmable api's.......could then use magnite cleardirect ip to get the webrtc(jitsi) audio/video messaging into the application using favascript /html5 of chrome and firefox?

then call all the app dev ops...use also the microfrontend registry of preconfigured objects and pretrained ai on such. guesses.

yeah, insiders just sold about 2k shares each to fulfill their obligations, but no mass panic.

heavy censoring on the yahoo board. not even to the rejection step, just blank.
but
"The WebRTC APIs essentially allow JavaScript to generate real-time video and voice streams that can be treated like standard video and audio objects in HTML5. This gives developers a chance to incorporate streaming media into their applications using nothing more than HTML5 capabilities that are already built into Chrome and Firefox browsers.
WebRTC and HTML5: An Overview - OnSIP
www.onsip.com/voip-resources/voip-fundamentals/webrtc-html5-overview
www.onsip.com/voip-resources/voip-fundamentals/webrtc-html5-overvie"
add this to multiple chat apps with programmable apis.....and maybe a way to do the in app audio /video and messaging ip in the iframe /google amp. with the amazon cloudfront? 8x8 cloud level...of which there appear to be different levels 1 2 3. depending on size?

heavy censoring on the yahoo board. not even to the rejection step, just blank.
but
"The WebRTC APIs essentially allow JavaScript to generate real-time video and voice streams that can be treated like standard video and audio objects in HTML5. This gives developers a chance to incorporate streaming media into their applications using nothing more than HTML5 capabilities that are already built into Chrome and Firefox browsers.
WebRTC and HTML5: An Overview - OnSIP
www.onsip.com/voip-resources/voip-fundamentals/webrtc-html5-overview
www.onsip.com/voip-resources/voip-fundamentals/webrtc-html5-overvie"
add this to multiple chat apps with programmable apis.....and maybe a way to do the in app audio /video and messaging ip in the iframe /google amp. with the amazon cloudfront? 8x8 cloud level...of which there appear to be different levels 1 2 3. depending on size?

What is difference between syntax and semantics in artificial intelligence?
What Is syntax AI?
Syntax - It refers to arranging words to make a sentence. It also involves determining the structural role of words in the sentence and in phrases. Semantics - It is concerned with the meaning of words and how to combine words into meaningful phrases and sentences.

natural language prompts and ability to translate and transcribe. etl........
now i wonder if a prompt is calling for the correct virtual server/virtual service to do you work?
one for financial, another for football team, another for medical, another for chemistry/physics/,
another for medical specialties like eye, ent, cardiac, endocrinology or across the board medicine?
another for dna sequencing/synthetic construction.
etc.

yahoo blocking 8x8 board . or censoring.
linked in msft copilot presentation. wow.
but it requires getting your business' unified communications .email/text/chat/calls/mail/etc.
they use llm, graph data, modified prompts.

yes, 8's prior ceo was spot on.
now thinking ibm and threats/homeland security work.
or power grid resiliency/maintenance
or nuclear power

https://www.linkedin.com/events/thefutureofwork-reinventingprod7038508574142074880/comments/

yahoo boards are censoring mightily today. obviously, a crooked way to have a discussion forum.

msft new work with ai at 11am et today.
8x8 platform is integrated with open ai.
8x8 voice for msft teams.
ai works with data ...all of the data would yield better ai, lest you miss something.

capped calls on 8x8 14million shares; ? to cover their shorts? priced 25 /share?
elliot bought citrix to merge with tibco; tibco and apache kafka....streaming realtime data..? some jitsi/8x8 and oracle which is less expressive streaming ip?
francisco loaned the money to 8x8;
francisco bought watson healthcare.
citrix used in many hospitals. problem is each app has to be opened individually? 8x8 multiple chat apps in an app could be huge
mgh/brigham group?

all of a sudden they're looking at historical data and finding more evidence of balloon incursions.

google maps and zerigo integration
edge detection and motion vector estimation
tumor mapping for radiologic or electromagnetic field treatment
rfid to label the complex polyhedron shapes
microfocus/msft azure partnership for ai on visual object transformation

examples of posts which were rejected by yahoo finance 8x8. why? obvious.?

google is down today because of rumored faulty information for a search on james webb telescope. the ai said james webb was the first to take pictures of exoplanets, but it was done earlier . However, a search of nasa and james webb disclosed that james webb would be taking/took pictures of exoplanets in never before seen wavelengths which would yield new information. so, not exactly right but not wrong. the meta data should have indicated what kind of pictures...new wavelength pictures. perhaps, like any other ml/ai , the more you use it, the better it gets?

microservices and generative ai.
X is the cloud native collaboration level of microservices?
it is cloud level......google amp, amazon cloudfront, 8x8 , wags.

vall-e: extract, make it into tokens(transform), load.....interesting, eh?

FUZE and apple. perhaps, apple sensors and FUZE connectivity in mobile client, with FUZE now becoming 8x8 Client/work for mobile? sensors collecting early warning data/metrics. combined with subsequent biomarkers/contract tracing and initiation of tlr7 system....analagous to looking for drone infiltration? guesses.

hp bought neoware and now poly. poly used 8x8 network. new partner of 8x8 hp.

hp partner of vmware and msft and cisco and intel and amd and google and ect.

vista equity and citrix which is to be integrated with tibco.

citrix client to msft os/pc's. for connection to server/services/and other organizations.

8x8 and cloud ip. including nodes and networked nodes.

connectivity via ethereum blockchain, merge proof of stake.

contracts at the client side determine the functionality/routing/etc.

data about corporate interactions analyzed in realtime.

sales force organization can be redone with significantly less effort than before. this would facilitate mergers and acquisitions and reorganizations.

we all remember 1-800 flowers, or some of us do.

adhoc working groups now becoming de rigueur in many organizations due to job, goals, employee attendance, etc.

amazon inspire ? could be made with 8x8 cloud level......amazon cloudfront and google amp and jitsi for streaming webrtc....

https://ubunlog.com/en/jitsi-an-excellent-open-source-multiplatform-video-conferencing-application/#Que_hay_de_nuevo_en_la_version_20

needham conf.....autobots for cc....overall, 8x8 people seemed almost reticent to be there talking about company. potentially related to corp layoffs in techs....roi improvements with 8x8 ip alluded to in a general way. more earnings with less revenue? no questions about outside forces interested in 8x8 buyout...but recent tech pr's and new ip software points to an ability to create vertical blockchains of productivity. and blockchains are now proof of stake. i wonder if 8x8 was using such for their virtual meetings ip?

Russian missiles into Polanda? Market reverses. 8x8 offers security and privacy of comms, which might be desirable in this climate.

blockchain merge/pos......5 dollars for twitter account?
more for additiional benefits.......like twitter pro? provisioned by jitsi as a service?
add it to google amp/browser and ad/no ad supported streaming. wags.

fuze card, a multi credit card/pass work manager.....hmm.....yah think it might use chat apps? as in multiple chat apps in an app with programmable apis? think it's covered by the singapore security certificate? wags?

yahoo posting still locked down, but.............:
Avg Vol (3 month) 3 2.36M
Avg Vol (10 day) 3 2.01M
Shares Outstanding 5 109.27M
Implied Shares Outstanding 6 N/A
Float 8 90.04M
% Held by Insiders 1 3.06%
% Held by Institutions 1 99.04%
Shares Short (Oct 14, 2022) 4 16.22M
Short Ratio (Oct 14, 2022) 4 8.29
Short % of Float (Oct 14, 2022) 4 25.23%
Short % of Shares Outstanding (Oct 14, 2022) 4 14.85%
Shares Short (prior month Sep 15, 2022) 4 13.92M
'

cointelegraph.com/news/google-launches-blockchain-node-engine-for-web-3-0-developersAccording to an Oct. 27 announcement, tech-giant Google is launching an in-house Blockchain Node Engine (BNE) based on Google Cloud. As a node-hosting service, BNE enables Web3 companies to relay transactions, deploy smart contracts, and read or write blockchain data directly on Google Cloud. Ethereum will be the first blockchain supported by BNE. As told by Google:

“Today, manually deploying a node is a time-intensive process that involves provisioning a compute instance, installing an Ethereum client, and waiting for the node to sync with the network. [...] Google Cloud’s BNE can make this process faster and easier by allowing developers to deploy a new node with a single operation and specify the desired region and network.”============

note..........relay transactions......ethereum will be the first supported by bne......ethereum merge essential brave browser, which has deal with 8x8 integration. google cloud ai deal with 8x8....oh.......so, bne would seem to be an 8x8 deal and vertical thing with 8x8 network. wags.

Got filled on my options now looking great

EGHT is looking great right now I wanted to get some 35 call options this morning and got side tracked oh well.

EGHT is looking great right now I wanted to get some 35 call options this morning and got side tracked oh well.